Merged from main + fixed elasticsearch_url

tests/common/mcp.py

@@ -244,8 +244,14 @@ def make_mcp_server(required_auth_token: str | None = None, tools: dict[str, Cal
     timeout = 2
     start_time = time.time()
 
-    server_url = f"http://localhost:{port}/sse"
-    logger.debug(f"Waiting for MCP server thread to start on port {port}")
+    # Determine the appropriate host for the server URL based on test environment
+    # - For library client and server mode: use localhost (both on same host)
+    # - For docker mode: use host.docker.internal (container needs to reach host)
+    import os
+
+    mcp_host = os.environ.get("LLAMA_STACK_TEST_MCP_HOST", "localhost")
+    server_url = f"http://{mcp_host}:{port}/sse"
+    logger.debug(f"Waiting for MCP server thread to start on port {port} (accessible via {mcp_host})")
 
     while time.time() - start_time < timeout:
         if server_thread.is_alive():

tests/external/llama-stack-api-weather/src/llama_stack_api_weather/weather.py

@@ -6,9 +6,7 @@
 
 from typing import Protocol
 
-from llama_stack.apis.version import LLAMA_STACK_API_V1
-from llama_stack.providers.datatypes import Api, ProviderSpec, RemoteProviderSpec
-from llama_stack.schema_utils import webmethod
+from llama_stack_api import LLAMA_STACK_API_V1, Api, ProviderSpec, RemoteProviderSpec, webmethod
 
 
 def available_providers() -> list[ProviderSpec]:

tests/integration/agents/test_openai_responses.py

@@ -516,3 +516,169 @@ def test_response_with_instructions(openai_client, client_with_models, text_mode
 
     # Verify instructions from previous response was not carried over to the next response
     assert response_with_instructions2.instructions == instructions2
+
+
+@pytest.mark.skip(reason="Tool calling is not reliable.")
+def test_max_tool_calls_with_function_tools(openai_client, client_with_models, text_model_id):
+    """Test handling of max_tool_calls with function tools in responses."""
+    if isinstance(client_with_models, LlamaStackAsLibraryClient):
+        pytest.skip("OpenAI responses are not supported when testing with library client yet.")
+
+    client = openai_client
+    max_tool_calls = 1
+
+    tools = [
+        {
+            "type": "function",
+            "name": "get_weather",
+            "description": "Get weather information for a specified location",
+            "parameters": {
+                "type": "object",
+                "properties": {
+                    "location": {
+                        "type": "string",
+                        "description": "The city name (e.g., 'New York', 'London')",
+                    },
+                },
+            },
+        },
+        {
+            "type": "function",
+            "name": "get_time",
+            "description": "Get current time for a specified location",
+            "parameters": {
+                "type": "object",
+                "properties": {
+                    "location": {
+                        "type": "string",
+                        "description": "The city name (e.g., 'New York', 'London')",
+                    },
+                },
+            },
+        },
+    ]
+
+    # First create a response that triggers function tools
+    response = client.responses.create(
+        model=text_model_id,
+        input="Can you tell me the weather in Paris and the current time?",
+        tools=tools,
+        stream=False,
+        max_tool_calls=max_tool_calls,
+    )
+
+    # Verify we got two function calls and that the max_tool_calls do not affect function tools
+    assert len(response.output) == 2
+    assert response.output[0].type == "function_call"
+    assert response.output[0].name == "get_weather"
+    assert response.output[0].status == "completed"
+    assert response.output[1].type == "function_call"
+    assert response.output[1].name == "get_time"
+    assert response.output[0].status == "completed"
+
+    # Verify we have a valid max_tool_calls field
+    assert response.max_tool_calls == max_tool_calls
+
+
+def test_max_tool_calls_invalid(openai_client, client_with_models, text_model_id):
+    """Test handling of invalid max_tool_calls in responses."""
+    if isinstance(client_with_models, LlamaStackAsLibraryClient):
+        pytest.skip("OpenAI responses are not supported when testing with library client yet.")
+
+    client = openai_client
+
+    input = "Search for today's top technology news."
+    invalid_max_tool_calls = 0
+    tools = [
+        {"type": "web_search"},
+    ]
+
+    # Create a response with an invalid max_tool_calls value i.e. 0
+    # Handle ValueError from LLS and BadRequestError from OpenAI client
+    with pytest.raises((ValueError, BadRequestError)) as excinfo:
+        client.responses.create(
+            model=text_model_id,
+            input=input,
+            tools=tools,
+            stream=False,
+            max_tool_calls=invalid_max_tool_calls,
+        )
+
+    error_message = str(excinfo.value)
+    assert f"Invalid max_tool_calls={invalid_max_tool_calls}; should be >= 1" in error_message, (
+        f"Expected error message about invalid max_tool_calls, got: {error_message}"
+    )
+
+
+def test_max_tool_calls_with_builtin_tools(openai_client, client_with_models, text_model_id):
+    """Test handling of max_tool_calls with built-in tools in responses."""
+    if isinstance(client_with_models, LlamaStackAsLibraryClient):
+        pytest.skip("OpenAI responses are not supported when testing with library client yet.")
+
+    client = openai_client
+
+    input = "Search for today's top technology and a positive news story. You MUST make exactly two separate web search calls."
+    max_tool_calls = [1, 5]
+    tools = [
+        {"type": "web_search"},
+    ]
+
+    # First create a response that triggers web_search tools without max_tool_calls
+    response = client.responses.create(
+        model=text_model_id,
+        input=input,
+        tools=tools,
+        stream=False,
+    )
+
+    # Verify we got two web search calls followed by a message
+    assert len(response.output) == 3
+    assert response.output[0].type == "web_search_call"
+    assert response.output[0].status == "completed"
+    assert response.output[1].type == "web_search_call"
+    assert response.output[1].status == "completed"
+    assert response.output[2].type == "message"
+    assert response.output[2].status == "completed"
+    assert response.output[2].role == "assistant"
+
+    # Next create a response that triggers web_search tools with max_tool_calls set to 1
+    response_2 = client.responses.create(
+        model=text_model_id,
+        input=input,
+        tools=tools,
+        stream=False,
+        max_tool_calls=max_tool_calls[0],
+    )
+
+    # Verify we got one web search tool call followed by a message
+    assert len(response_2.output) == 2
+    assert response_2.output[0].type == "web_search_call"
+    assert response_2.output[0].status == "completed"
+    assert response_2.output[1].type == "message"
+    assert response_2.output[1].status == "completed"
+    assert response_2.output[1].role == "assistant"
+
+    # Verify we have a valid max_tool_calls field
+    assert response_2.max_tool_calls == max_tool_calls[0]
+
+    # Finally create a response that triggers web_search tools with max_tool_calls set to 5
+    response_3 = client.responses.create(
+        model=text_model_id,
+        input=input,
+        tools=tools,
+        stream=False,
+        max_tool_calls=max_tool_calls[1],
+    )
+
+    # Verify we got two web search calls followed by a message
+    assert len(response_3.output) == 3
+    assert response_3.output[0].type == "web_search_call"
+    assert response_3.output[0].status == "completed"
+    assert response_3.output[1].type == "web_search_call"
+    assert response_3.output[1].status == "completed"
+    assert response_3.output[2].type == "message"
+    assert response_3.output[2].status == "completed"
+    assert response_3.output[2].role == "assistant"
+
+    # Verify we have a valid max_tool_calls field
+    assert response_3.max_tool_calls == max_tool_calls[1]

tests/integration/batches/conftest.py

@@ -14,7 +14,7 @@ from io import BytesIO
 
 import pytest
 
-from llama_stack.apis.files import OpenAIFilePurpose
+from llama_stack_api import OpenAIFilePurpose
 
 
 class BatchHelper:

tests/integration/ci_matrix.json

@@ -1,6 +1,7 @@
 {
   "default": [
     {"suite": "base", "setup": "ollama"},
+    {"suite": "base", "setup": "ollama-postgres", "allowed_clients": ["server"], "stack_config": "server:ci-tests::run-with-postgres-store.yaml"},
     {"suite": "vision", "setup": "ollama-vision"},
     {"suite": "responses", "setup": "gpt"},
     {"suite": "base-vllm-subset", "setup": "vllm"}

tests/integration/files/test_files.py

@@ -10,8 +10,8 @@ from unittest.mock import patch
 import pytest
 import requests
 
-from llama_stack.apis.files import OpenAIFilePurpose
 from llama_stack.core.datatypes import User
+from llama_stack_api import OpenAIFilePurpose
 
 purpose = OpenAIFilePurpose.ASSISTANTS
 
@@ -175,7 +175,7 @@ def test_expires_after_requests(openai_client):
 
 
 @pytest.mark.xfail(message="User isolation broken for current providers, must be fixed.")
-@patch("llama_stack.providers.utils.sqlstore.authorized_sqlstore.get_authenticated_user")
+@patch("llama_stack.core.storage.sqlstore.authorized_sqlstore.get_authenticated_user")
 def test_files_authentication_isolation(mock_get_authenticated_user, llama_stack_client):
     """Test that users can only access their own files."""
     from llama_stack_client import NotFoundError
@@ -275,7 +275,7 @@ def test_files_authentication_isolation(mock_get_authenticated_user, llama_stack
         raise e
 
 
-@patch("llama_stack.providers.utils.sqlstore.authorized_sqlstore.get_authenticated_user")
+@patch("llama_stack.core.storage.sqlstore.authorized_sqlstore.get_authenticated_user")
 def test_files_authentication_shared_attributes(
     mock_get_authenticated_user, llama_stack_client, provider_type_is_openai
 ):
@@ -335,7 +335,7 @@ def test_files_authentication_shared_attributes(
         raise e
 
 
-@patch("llama_stack.providers.utils.sqlstore.authorized_sqlstore.get_authenticated_user")
+@patch("llama_stack.core.storage.sqlstore.authorized_sqlstore.get_authenticated_user")
 def test_files_authentication_anonymous_access(
     mock_get_authenticated_user, llama_stack_client, provider_type_is_openai
 ):

tests/integration/fixtures/common.py

@@ -233,10 +233,21 @@ def instantiate_llama_stack_client(session):
         raise ValueError("You must specify either --stack-config or LLAMA_STACK_CONFIG")
 
     # Handle server:<config_name> format or server:<config_name>:<port>
+    # Also handles server:<distro>::<run_file.yaml> format
     if config.startswith("server:"):
-        parts = config.split(":")
-        config_name = parts[1]
-        port = int(parts[2]) if len(parts) > 2 else int(os.environ.get("LLAMA_STACK_PORT", DEFAULT_PORT))
+        # Strip the "server:" prefix first
+        config_part = config[7:]  # len("server:") == 7
+
+        # Check for :: (distro::runfile format)
+        if "::" in config_part:
+            config_name = config_part
+            port = int(os.environ.get("LLAMA_STACK_PORT", DEFAULT_PORT))
+        else:
+            # Single colon format: either <name> or <name>:<port>
+            parts = config_part.split(":")
+            config_name = parts[0]
+            port = int(parts[1]) if len(parts) > 1 else int(os.environ.get("LLAMA_STACK_PORT", DEFAULT_PORT))
+
         base_url = f"http://localhost:{port}"
 
         force_restart = os.environ.get("LLAMA_STACK_TEST_FORCE_SERVER_RESTART") == "1"
@@ -323,7 +334,13 @@ def require_server(llama_stack_client):
 @pytest.fixture(scope="session")
 def openai_client(llama_stack_client, require_server):
     base_url = f"{llama_stack_client.base_url}/v1"
-    return OpenAI(base_url=base_url, api_key="fake")
+    client = OpenAI(base_url=base_url, api_key="fake", max_retries=0, timeout=30.0)
+    yield client
+    # Cleanup: close HTTP connections
+    try:
+        client.close()
+    except Exception:
+        pass
 
 
 @pytest.fixture(params=["openai_client", "client_with_models"])

tests/integration/inference/test_openai_completion.py

@@ -54,6 +54,7 @@ def skip_if_model_doesnt_support_openai_completion(client_with_models, model_id)
         # {"error":{"message":"Unknown request URL: GET /openai/v1/completions. Please check the URL for typos,
         # or see the docs at https://console.groq.com/docs/","type":"invalid_request_error","code":"unknown_url"}}
         "remote::groq",
+        "remote::oci",
         "remote::gemini",  # https://generativelanguage.googleapis.com/v1beta/openai/completions -> 404
         "remote::anthropic",  # at least claude-3-{5,7}-{haiku,sonnet}-* / claude-{sonnet,opus}-4-* are not supported
         "remote::azure",  # {'error': {'code': 'OperationNotSupported', 'message': 'The completion operation

tests/integration/inference/test_openai_embeddings.py

@@ -138,6 +138,7 @@ def skip_if_model_doesnt_support_openai_embeddings(client, model_id):
         "remote::runpod",
         "remote::sambanova",
         "remote::tgi",
+        "remote::oci",
     ):
         pytest.skip(f"Model {model_id} hosted by {provider.provider_type} doesn't support OpenAI embeddings.")
 

tests/integration/inference/test_provider_data_routing.py

@@ -16,15 +16,15 @@ from unittest.mock import AsyncMock, patch
 
 import pytest
 
-from llama_stack.apis.datatypes import Api
-from llama_stack.apis.inference.inference import (
+from llama_stack.core.library_client import LlamaStackAsLibraryClient
+from llama_stack.core.telemetry.telemetry import MetricEvent
+from llama_stack_api import (
+    Api,
     OpenAIAssistantMessageParam,
     OpenAIChatCompletion,
     OpenAIChatCompletionUsage,
     OpenAIChoice,
 )
-from llama_stack.core.library_client import LlamaStackAsLibraryClient
-from llama_stack.core.telemetry.telemetry import MetricEvent
 
 
 class OpenAIChatCompletionWithMetrics(OpenAIChatCompletion):

tests/integration/inference/test_tools_with_schemas.py

@@ -9,8 +9,6 @@ Integration tests for inference/chat completion with JSON Schema-based tools.
 Tests that tools pass through correctly to various LLM providers.
 """
 
-import json
-
 import pytest
 
 from llama_stack.core.library_client import LlamaStackAsLibraryClient
@@ -193,15 +191,11 @@ class TestMCPToolsInChatCompletion:
             mcp_endpoint=dict(uri=uri),
         )
 
-        provider_data = {"mcp_headers": {uri: {"Authorization": f"Bearer {AUTH_TOKEN}"}}}
-        auth_headers = {
-            "X-LlamaStack-Provider-Data": json.dumps(provider_data),
-        }
-
+        # Use the dedicated authorization parameter
         # Get the tools from MCP
         tools_response = llama_stack_client.tool_runtime.list_tools(
             tool_group_id=test_toolgroup_id,
-            extra_headers=auth_headers,
+            authorization=AUTH_TOKEN,
         )
 
         # Convert to OpenAI format for inference

tests/integration/post_training/test_post_training.py

@@ -10,7 +10,8 @@ import uuid
 
 import pytest
 
-from llama_stack.apis.post_training import (
+from llama_stack.log import get_logger
+from llama_stack_api import (
     DataConfig,
     DatasetFormat,
     DPOAlignmentConfig,
@@ -18,7 +19,6 @@ from llama_stack.apis.post_training import (
     LoraFinetuningConfig,
     TrainingConfig,
 )
-from llama_stack.log import get_logger
 
 # Configure logging
 logger = get_logger(name=__name__, category="post_training")

tests/integration/providers/utils/sqlstore/test_authorized_sqlstore.py

@@ -13,14 +13,14 @@ import pytest
 from llama_stack.core.access_control.access_control import default_policy
 from llama_stack.core.datatypes import User
 from llama_stack.core.storage.datatypes import SqlStoreReference
-from llama_stack.providers.utils.sqlstore.api import ColumnType
-from llama_stack.providers.utils.sqlstore.authorized_sqlstore import AuthorizedSqlStore
-from llama_stack.providers.utils.sqlstore.sqlstore import (
+from llama_stack.core.storage.sqlstore.authorized_sqlstore import AuthorizedSqlStore
+from llama_stack.core.storage.sqlstore.sqlstore import (
     PostgresSqlStoreConfig,
     SqliteSqlStoreConfig,
     register_sqlstore_backends,
     sqlstore_impl,
 )
+from llama_stack_api.internal.sqlstore import ColumnType
 
 
 def get_postgres_config():
@@ -96,7 +96,7 @@ async def cleanup_records(sql_store, table_name, record_ids):
 
 
 @pytest.mark.parametrize("backend_config", BACKEND_CONFIGS)
-@patch("llama_stack.providers.utils.sqlstore.authorized_sqlstore.get_authenticated_user")
+@patch("llama_stack.core.storage.sqlstore.authorized_sqlstore.get_authenticated_user")
 async def test_authorized_store_attributes(mock_get_authenticated_user, authorized_store, request):
     """Test that JSON column comparisons work correctly for both PostgreSQL and SQLite"""
     backend_name = request.node.callspec.id
@@ -190,7 +190,7 @@ async def test_authorized_store_attributes(mock_get_authenticated_user, authoriz
 
 
 @pytest.mark.parametrize("backend_config", BACKEND_CONFIGS)
-@patch("llama_stack.providers.utils.sqlstore.authorized_sqlstore.get_authenticated_user")
+@patch("llama_stack.core.storage.sqlstore.authorized_sqlstore.get_authenticated_user")
 async def test_user_ownership_policy(mock_get_authenticated_user, authorized_store, request):
     """Test that 'user is owner' policies work correctly with record ownership"""
     from llama_stack.core.access_control.datatypes import AccessRule, Action, Scope

tests/integration/recordings/README.md

@@ -2,6 +2,10 @@
 
 This directory contains recorded inference API responses used for deterministic testing without requiring live API access.
 
+For more information, see the
+[docs](https://llamastack.github.io/docs/contributing/testing/record-replay).
+This README provides more technical information.
+
 ## Structure
 
 - `responses/` - JSON files containing request/response pairs for inference operations

tests/integration/responses/conftest.py

@@ -0,0 +1,17 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+import pytest
+
+from llama_stack.core.library_client import LlamaStackAsLibraryClient
+
+
+@pytest.fixture
+def responses_client(compat_client):
+    """Provide a client for responses tests, skipping library client mode."""
+    if isinstance(compat_client, LlamaStackAsLibraryClient):
+        pytest.skip("Responses API tests are not supported in library client mode")
+    return compat_client

tests/integration/responses/fixtures/fixtures.py

@@ -115,7 +115,15 @@ def openai_client(base_url, api_key, provider):
         client = LlamaStackAsLibraryClient(config, skip_logger_removal=True)
         return client
 
-    return OpenAI(
+    client = OpenAI(
         base_url=base_url,
         api_key=api_key,
+        max_retries=0,
+        timeout=30.0,
     )
+    yield client
+    # Cleanup: close HTTP connections
+    try:
+        client.close()
+    except Exception:
+        pass

tests/integration/responses/recordings/0a4aca0cd075369aaf6133ee82d9d940455cb083c0fd1330c666a12d74df6f89.json

@@ -0,0 +1,549 @@
+{
+  "test_id": "tests/integration/responses/test_tool_responses.py::test_response_streaming_multi_turn_tool_execution[openai_client-txt=openai/gpt-4o-experiment_analysis_streaming]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "I need a complete analysis: First, get the experiment ID for 'chemical_reaction', then get the results for that experiment, and tell me if the yield was above 80%. Return only one tool call per step.  Please stream your analysis process."
+        },
+        {
+          "role": "assistant",
+          "content": "",
+          "tool_calls": [
+            {
+              "index": 0,
+              "id": "call_Q9Gcxub7UbQsxJWVkiy4FETr",
+              "type": "function",
+              "function": {
+                "name": "get_experiment_id",
+                "arguments": "{\"experiment_name\":\"chemical_reaction\"}"
+              }
+            }
+          ]
+        },
+        {
+          "role": "tool",
+          "tool_call_id": "call_Q9Gcxub7UbQsxJWVkiy4FETr",
+          "content": [
+            {
+              "type": "text",
+              "text": "exp_003"
+            }
+          ]
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      },
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "get_user_id",
+            "description": "\n        Get the user ID for a given username. This ID is needed for other operations.\n\n        :param username: The username to look up\n        :return: The user ID for the username\n        ",
+            "parameters": {
+              "properties": {
+                "username": {
+                  "title": "Username",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "username"
+              ],
+              "title": "get_user_idArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_user_permissions",
+            "description": "\n        Get the permissions for a user ID. Requires a valid user ID from get_user_id.\n\n        :param user_id: The user ID to check permissions for\n        :return: The permissions for the user\n        ",
+            "parameters": {
+              "properties": {
+                "user_id": {
+                  "title": "User Id",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "user_id"
+              ],
+              "title": "get_user_permissionsArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "check_file_access",
+            "description": "\n        Check if a user can access a specific file. Requires a valid user ID.\n\n        :param user_id: The user ID to check access for\n        :param filename: The filename to check access to\n        :return: Whether the user can access the file (yes/no)\n        ",
+            "parameters": {
+              "properties": {
+                "user_id": {
+                  "title": "User Id",
+                  "type": "string"
+                },
+                "filename": {
+                  "title": "Filename",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "user_id",
+                "filename"
+              ],
+              "title": "check_file_accessArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_experiment_id",
+            "description": "\n        Get the experiment ID for a given experiment name. This ID is needed to get results.\n\n        :param experiment_name: The name of the experiment\n        :return: The experiment ID\n        ",
+            "parameters": {
+              "properties": {
+                "experiment_name": {
+                  "title": "Experiment Name",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "experiment_name"
+              ],
+              "title": "get_experiment_idArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_experiment_results",
+            "description": "\n        Get the results for an experiment ID. Requires a valid experiment ID from get_experiment_id.\n\n        :param experiment_id: The experiment ID to get results for\n        :return: The experiment results\n        ",
+            "parameters": {
+              "properties": {
+                "experiment_id": {
+                  "title": "Experiment Id",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "experiment_id"
+              ],
+              "title": "get_experiment_resultsArguments",
+              "type": "object"
+            }
+          }
+        }
+      ]
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-0a4aca0cd075",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": "call_yTMuQEKu7x115q8XvhqelRub",
+                    "function": {
+                      "arguments": "",
+                      "name": "get_experiment_results"
+                    },
+                    "type": "function"
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "9CSOZwfG5M7nid"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-0a4aca0cd075",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "{\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Wss"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-0a4aca0cd075",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "experiment",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "5AmVsa0S6NBy"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-0a4aca0cd075",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "_id",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "2Sf"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-0a4aca0cd075",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\":\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "z"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-0a4aca0cd075",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "exp",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "leu"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-0a4aca0cd075",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "_",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "omxpR"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-0a4aca0cd075",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "003",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "kW6"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-0a4aca0cd075",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\"}",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Zm6"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-0a4aca0cd075",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": "tool_calls",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "aXvC"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-0a4aca0cd075",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": {
+            "completion_tokens": 19,
+            "prompt_tokens": 457,
+            "total_tokens": 476,
+            "completion_tokens_details": {
+              "accepted_prediction_tokens": 0,
+              "audio_tokens": 0,
+              "reasoning_tokens": 0,
+              "rejected_prediction_tokens": 0
+            },
+            "prompt_tokens_details": {
+              "audio_tokens": 0,
+              "cached_tokens": 0
+            }
+          },
+          "obfuscation": "s13YHOCCaCDcJ"
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/recordings/2bd4c8dc08b3ee3ffce696864f0bd9f35d82223c7d1cab613ab2e818d79d6f9b.json

@@ -0,0 +1,295 @@
+{
+  "test_id": "tests/integration/responses/test_tool_responses.py::test_response_non_streaming_multi_turn_tool_execution[openai_client-txt=openai/gpt-4o-user_file_access_check]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "I need to check if user 'alice' can access the file 'document.txt'. First, get alice's user ID, then check if that user ID can access the file 'document.txt'. Do this as a series of steps, where each step is a separate message. Return only one tool call per step. Summarize the final result with a single 'yes' or 'no' response."
+        },
+        {
+          "role": "assistant",
+          "content": "",
+          "tool_calls": [
+            {
+              "index": 0,
+              "id": "call_EsVvmBUqtJb42kNkYnK19QkJ",
+              "type": "function",
+              "function": {
+                "name": "get_user_id",
+                "arguments": "{\"username\":\"alice\"}"
+              }
+            }
+          ]
+        },
+        {
+          "role": "tool",
+          "tool_call_id": "call_EsVvmBUqtJb42kNkYnK19QkJ",
+          "content": [
+            {
+              "type": "text",
+              "text": "user_12345"
+            }
+          ]
+        },
+        {
+          "role": "assistant",
+          "content": "",
+          "tool_calls": [
+            {
+              "index": 0,
+              "id": "call_kCmSE8ORKfQoiEsW2UCYr5Sh",
+              "type": "function",
+              "function": {
+                "name": "check_file_access",
+                "arguments": "{\"user_id\":\"user_12345\",\"filename\":\"document.txt\"}"
+              }
+            }
+          ]
+        },
+        {
+          "role": "tool",
+          "tool_call_id": "call_kCmSE8ORKfQoiEsW2UCYr5Sh",
+          "content": [
+            {
+              "type": "text",
+              "text": "yes"
+            }
+          ]
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      },
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "get_user_id",
+            "description": "\n        Get the user ID for a given username. This ID is needed for other operations.\n\n        :param username: The username to look up\n        :return: The user ID for the username\n        ",
+            "parameters": {
+              "properties": {
+                "username": {
+                  "title": "Username",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "username"
+              ],
+              "title": "get_user_idArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_user_permissions",
+            "description": "\n        Get the permissions for a user ID. Requires a valid user ID from get_user_id.\n\n        :param user_id: The user ID to check permissions for\n        :return: The permissions for the user\n        ",
+            "parameters": {
+              "properties": {
+                "user_id": {
+                  "title": "User Id",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "user_id"
+              ],
+              "title": "get_user_permissionsArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "check_file_access",
+            "description": "\n        Check if a user can access a specific file. Requires a valid user ID.\n\n        :param user_id: The user ID to check access for\n        :param filename: The filename to check access to\n        :return: Whether the user can access the file (yes/no)\n        ",
+            "parameters": {
+              "properties": {
+                "user_id": {
+                  "title": "User Id",
+                  "type": "string"
+                },
+                "filename": {
+                  "title": "Filename",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "user_id",
+                "filename"
+              ],
+              "title": "check_file_accessArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_experiment_id",
+            "description": "\n        Get the experiment ID for a given experiment name. This ID is needed to get results.\n\n        :param experiment_name: The name of the experiment\n        :return: The experiment ID\n        ",
+            "parameters": {
+              "properties": {
+                "experiment_name": {
+                  "title": "Experiment Name",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "experiment_name"
+              ],
+              "title": "get_experiment_idArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_experiment_results",
+            "description": "\n        Get the results for an experiment ID. Requires a valid experiment ID from get_experiment_id.\n\n        :param experiment_id: The experiment ID to get results for\n        :return: The experiment results\n        ",
+            "parameters": {
+              "properties": {
+                "experiment_id": {
+                  "title": "Experiment Id",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "experiment_id"
+              ],
+              "title": "get_experiment_resultsArguments",
+              "type": "object"
+            }
+          }
+        }
+      ]
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-2bd4c8dc08b3",
+          "choices": [
+            {
+              "delta": {
+                "content": "",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "UxHf8fChwO3CUY"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-2bd4c8dc08b3",
+          "choices": [
+            {
+              "delta": {
+                "content": "yes",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "GOexNEhopELIg"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-2bd4c8dc08b3",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": "stop",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "O41d8hC8zD"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-2bd4c8dc08b3",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": {
+            "completion_tokens": 2,
+            "prompt_tokens": 516,
+            "total_tokens": 518,
+            "completion_tokens_details": {
+              "accepted_prediction_tokens": 0,
+              "audio_tokens": 0,
+              "reasoning_tokens": 0,
+              "rejected_prediction_tokens": 0
+            },
+            "prompt_tokens_details": {
+              "audio_tokens": 0,
+              "cached_tokens": 0
+            }
+          },
+          "obfuscation": "9VQklZAZMYAfa0"
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/recordings/2ed23a4289840f93202f94e7e7027118869d34d768ad87ba072e92e8a43a52f2.json

@@ -0,0 +1,833 @@
+{
+  "test_id": "tests/integration/responses/test_tool_responses.py::test_response_streaming_multi_turn_tool_execution[openai_client-txt=openai/gpt-4o-user_permissions_workflow]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "Help me with this security check: First, get the user ID for 'charlie', then get the permissions for that user ID, and finally check if that user can access 'secret_file.txt'. Stream your progress as you work through each step. Return only one tool call per step. Summarize the final result with a single 'yes' or 'no' response."
+        },
+        {
+          "role": "assistant",
+          "content": "",
+          "tool_calls": [
+            {
+              "index": 0,
+              "id": "call_fsxGbKmceUbLSXCe4sx9WLXO",
+              "type": "function",
+              "function": {
+                "name": "get_user_id",
+                "arguments": "{\"username\":\"charlie\"}"
+              }
+            }
+          ]
+        },
+        {
+          "role": "tool",
+          "tool_call_id": "call_fsxGbKmceUbLSXCe4sx9WLXO",
+          "content": [
+            {
+              "type": "text",
+              "text": "user_11111"
+            }
+          ]
+        },
+        {
+          "role": "assistant",
+          "content": "",
+          "tool_calls": [
+            {
+              "index": 0,
+              "id": "call_moRBxqnBJ48EWTSEoQ1llgib",
+              "type": "function",
+              "function": {
+                "name": "get_user_permissions",
+                "arguments": "{\"user_id\":\"user_11111\"}"
+              }
+            }
+          ]
+        },
+        {
+          "role": "tool",
+          "tool_call_id": "call_moRBxqnBJ48EWTSEoQ1llgib",
+          "content": [
+            {
+              "type": "text",
+              "text": "admin"
+            }
+          ]
+        },
+        {
+          "role": "assistant",
+          "content": "",
+          "tool_calls": [
+            {
+              "index": 0,
+              "id": "call_ybUqAP9oQn3rwQqVdOLs5Wb4",
+              "type": "function",
+              "function": {
+                "name": "check_file_access",
+                "arguments": "{\"user_id\":\"user_11111\",\"filename\":\"secret_file.txt\"}"
+              }
+            }
+          ]
+        },
+        {
+          "role": "tool",
+          "tool_call_id": "call_ybUqAP9oQn3rwQqVdOLs5Wb4",
+          "content": [
+            {
+              "type": "text",
+              "text": "no"
+            }
+          ]
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      },
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "get_user_id",
+            "description": "\n        Get the user ID for a given username. This ID is needed for other operations.\n\n        :param username: The username to look up\n        :return: The user ID for the username\n        ",
+            "parameters": {
+              "properties": {
+                "username": {
+                  "title": "Username",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "username"
+              ],
+              "title": "get_user_idArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_user_permissions",
+            "description": "\n        Get the permissions for a user ID. Requires a valid user ID from get_user_id.\n\n        :param user_id: The user ID to check permissions for\n        :return: The permissions for the user\n        ",
+            "parameters": {
+              "properties": {
+                "user_id": {
+                  "title": "User Id",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "user_id"
+              ],
+              "title": "get_user_permissionsArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "check_file_access",
+            "description": "\n        Check if a user can access a specific file. Requires a valid user ID.\n\n        :param user_id: The user ID to check access for\n        :param filename: The filename to check access to\n        :return: Whether the user can access the file (yes/no)\n        ",
+            "parameters": {
+              "properties": {
+                "user_id": {
+                  "title": "User Id",
+                  "type": "string"
+                },
+                "filename": {
+                  "title": "Filename",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "user_id",
+                "filename"
+              ],
+              "title": "check_file_accessArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_experiment_id",
+            "description": "\n        Get the experiment ID for a given experiment name. This ID is needed to get results.\n\n        :param experiment_name: The name of the experiment\n        :return: The experiment ID\n        ",
+            "parameters": {
+              "properties": {
+                "experiment_name": {
+                  "title": "Experiment Name",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "experiment_name"
+              ],
+              "title": "get_experiment_idArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_experiment_results",
+            "description": "\n        Get the results for an experiment ID. Requires a valid experiment ID from get_experiment_id.\n\n        :param experiment_id: The experiment ID to get results for\n        :return: The experiment results\n        ",
+            "parameters": {
+              "properties": {
+                "experiment_id": {
+                  "title": "Experiment Id",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "experiment_id"
+              ],
+              "title": "get_experiment_resultsArguments",
+              "type": "object"
+            }
+          }
+        }
+      ]
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-2ed23a428984",
+          "choices": [
+            {
+              "delta": {
+                "content": "",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "WLGSIGDbuImIc2"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-2ed23a428984",
+          "choices": [
+            {
+              "delta": {
+                "content": "The",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "tOPrT8GpCzqCn"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-2ed23a428984",
+          "choices": [
+            {
+              "delta": {
+                "content": " user",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "ViOvVDT7owF"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-2ed23a428984",
+          "choices": [
+            {
+              "delta": {
+                "content": " '",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "EkiYJGYtRb2KCr"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-2ed23a428984",
+          "choices": [
+            {
+              "delta": {
+                "content": "char",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "ioC2G58DuWTx"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-2ed23a428984",
+          "choices": [
+            {
+              "delta": {
+                "content": "lie",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "A5rxByl55APwi"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-2ed23a428984",
+          "choices": [
+            {
+              "delta": {
+                "content": "'",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "kmDNWRqOyy2r3ST"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-2ed23a428984",
+          "choices": [
+            {
+              "delta": {
+                "content": " cannot",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "JHGD4XKFC"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-2ed23a428984",
+          "choices": [
+            {
+              "delta": {
+                "content": " access",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "6IPkFhs93"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-2ed23a428984",
+          "choices": [
+            {
+              "delta": {
+                "content": " '",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "LGHjKnVq2lF1DS"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-2ed23a428984",
+          "choices": [
+            {
+              "delta": {
+                "content": "secret",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "1nGoXVjnK0"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-2ed23a428984",
+          "choices": [
+            {
+              "delta": {
+                "content": "_file",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "OeR7YlvZQLa"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-2ed23a428984",
+          "choices": [
+            {
+              "delta": {
+                "content": ".txt",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "yLKHaSgjE64R"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-2ed23a428984",
+          "choices": [
+            {
+              "delta": {
+                "content": "'.",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "waZY1Js7DPWtoN"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-2ed23a428984",
+          "choices": [
+            {
+              "delta": {
+                "content": " The",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "km3Gr5HspErW"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-2ed23a428984",
+          "choices": [
+            {
+              "delta": {
+                "content": " final",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Mvzf8AUstX"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-2ed23a428984",
+          "choices": [
+            {
+              "delta": {
+                "content": " result",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "660CrCPne"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-2ed23a428984",
+          "choices": [
+            {
+              "delta": {
+                "content": " is",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "lq7NyKvIo8UEO"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-2ed23a428984",
+          "choices": [
+            {
+              "delta": {
+                "content": ":",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "qjIz07y1RQsKqTo"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-2ed23a428984",
+          "choices": [
+            {
+              "delta": {
+                "content": " no",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "xhcVwxM4RaQcN"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-2ed23a428984",
+          "choices": [
+            {
+              "delta": {
+                "content": ".",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "dPxBJZ3WUesIy8T"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-2ed23a428984",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": "stop",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Z9wFfcEaK2"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-2ed23a428984",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": {
+            "completion_tokens": 21,
+            "prompt_tokens": 542,
+            "total_tokens": 563,
+            "completion_tokens_details": {
+              "accepted_prediction_tokens": 0,
+              "audio_tokens": 0,
+              "reasoning_tokens": 0,
+              "rejected_prediction_tokens": 0
+            },
+            "prompt_tokens_details": {
+              "audio_tokens": 0,
+              "cached_tokens": 0
+            }
+          },
+          "obfuscation": "fSoZk1lrb3nJt"
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/recordings/3177a984c900c2bdc2785b502bded6791b1054ce0f36e967eb3793b5608344f3.json

@@ -0,0 +1,759 @@
+{
+  "test_id": "tests/integration/responses/test_tool_responses.py::test_response_mcp_tool_approval[openai_client-txt=openai/gpt-4o-True-boiling_point_tool]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "What is the boiling point of myawesomeliquid in Celsius?"
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      },
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "greet_everyone",
+            "parameters": {
+              "properties": {
+                "url": {
+                  "title": "Url",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "url"
+              ],
+              "title": "greet_everyoneArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_boiling_point",
+            "description": "\n        Returns the boiling point of a liquid in Celsius or Fahrenheit.\n\n        :param liquid_name: The name of the liquid\n        :param celsius: Whether to return the boiling point in Celsius\n        :return: The boiling point of the liquid in Celcius or Fahrenheit\n        ",
+            "parameters": {
+              "properties": {
+                "liquid_name": {
+                  "title": "Liquid Name",
+                  "type": "string"
+                },
+                "celsius": {
+                  "default": true,
+                  "title": "Celsius",
+                  "type": "boolean"
+                }
+              },
+              "required": [
+                "liquid_name"
+              ],
+              "title": "get_boiling_pointArguments",
+              "type": "object"
+            }
+          }
+        }
+      ]
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-3177a984c900",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": "call_bL84OWNnE1s75GJEqGLAK35W",
+                    "function": {
+                      "arguments": "",
+                      "name": "get_boiling_point"
+                    },
+                    "type": "function"
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "ptE"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-3177a984c900",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "{\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "UEV"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-3177a984c900",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "li",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "hMko"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-3177a984c900",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "quid",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "nr"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-3177a984c900",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "_name",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "x"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-3177a984c900",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\":\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "D"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-3177a984c900",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "my",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "aLLC"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-3177a984c900",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "aw",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "EZdr"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-3177a984c900",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "esom",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "yV"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-3177a984c900",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "eli",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "0bj"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-3177a984c900",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "quid",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "5J"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-3177a984c900",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\",\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "z"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-3177a984c900",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "c",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "7dZEY"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-3177a984c900",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "elsius",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": ""
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-3177a984c900",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\":",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "AqP"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-3177a984c900",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "true",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "X8"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-3177a984c900",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "}",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "oa7h2"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-3177a984c900",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": "tool_calls",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "1Is8"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-3177a984c900",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": {
+            "completion_tokens": 27,
+            "prompt_tokens": 156,
+            "total_tokens": 183,
+            "completion_tokens_details": {
+              "accepted_prediction_tokens": 0,
+              "audio_tokens": 0,
+              "reasoning_tokens": 0,
+              "rejected_prediction_tokens": 0
+            },
+            "prompt_tokens_details": {
+              "audio_tokens": 0,
+              "cached_tokens": 0
+            }
+          },
+          "obfuscation": "DfwHMdbjUVww7"
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/recordings/318c5361647df0245c074cd2c7d6f50e862aeddbbeaeb256ef1add34de7c1dc8.json

@@ -0,0 +1,549 @@
+{
+  "test_id": "tests/integration/responses/test_tool_responses.py::test_response_non_streaming_multi_turn_tool_execution[openai_client-txt=openai/gpt-4o-experiment_results_lookup]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "I need to get the results for the 'boiling_point' experiment. First, get the experiment ID for 'boiling_point', then use that ID to get the experiment results. Tell me the boiling point in Celsius."
+        },
+        {
+          "role": "assistant",
+          "content": "",
+          "tool_calls": [
+            {
+              "index": 0,
+              "id": "call_dZwjBxH3aTRhnaS0bJVPqRcz",
+              "type": "function",
+              "function": {
+                "name": "get_experiment_id",
+                "arguments": "{\"experiment_name\":\"boiling_point\"}"
+              }
+            }
+          ]
+        },
+        {
+          "role": "tool",
+          "tool_call_id": "call_dZwjBxH3aTRhnaS0bJVPqRcz",
+          "content": [
+            {
+              "type": "text",
+              "text": "exp_004"
+            }
+          ]
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      },
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "get_user_id",
+            "description": "\n        Get the user ID for a given username. This ID is needed for other operations.\n\n        :param username: The username to look up\n        :return: The user ID for the username\n        ",
+            "parameters": {
+              "properties": {
+                "username": {
+                  "title": "Username",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "username"
+              ],
+              "title": "get_user_idArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_user_permissions",
+            "description": "\n        Get the permissions for a user ID. Requires a valid user ID from get_user_id.\n\n        :param user_id: The user ID to check permissions for\n        :return: The permissions for the user\n        ",
+            "parameters": {
+              "properties": {
+                "user_id": {
+                  "title": "User Id",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "user_id"
+              ],
+              "title": "get_user_permissionsArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "check_file_access",
+            "description": "\n        Check if a user can access a specific file. Requires a valid user ID.\n\n        :param user_id: The user ID to check access for\n        :param filename: The filename to check access to\n        :return: Whether the user can access the file (yes/no)\n        ",
+            "parameters": {
+              "properties": {
+                "user_id": {
+                  "title": "User Id",
+                  "type": "string"
+                },
+                "filename": {
+                  "title": "Filename",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "user_id",
+                "filename"
+              ],
+              "title": "check_file_accessArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_experiment_id",
+            "description": "\n        Get the experiment ID for a given experiment name. This ID is needed to get results.\n\n        :param experiment_name: The name of the experiment\n        :return: The experiment ID\n        ",
+            "parameters": {
+              "properties": {
+                "experiment_name": {
+                  "title": "Experiment Name",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "experiment_name"
+              ],
+              "title": "get_experiment_idArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_experiment_results",
+            "description": "\n        Get the results for an experiment ID. Requires a valid experiment ID from get_experiment_id.\n\n        :param experiment_id: The experiment ID to get results for\n        :return: The experiment results\n        ",
+            "parameters": {
+              "properties": {
+                "experiment_id": {
+                  "title": "Experiment Id",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "experiment_id"
+              ],
+              "title": "get_experiment_resultsArguments",
+              "type": "object"
+            }
+          }
+        }
+      ]
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-318c5361647d",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": "call_skNUKbERbtdoADH834U9OE91",
+                    "function": {
+                      "arguments": "",
+                      "name": "get_experiment_results"
+                    },
+                    "type": "function"
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "5aHvu2xes6Amy8"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-318c5361647d",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "{\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "9HQ"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-318c5361647d",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "experiment",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "ckAh5OXg9JIe"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-318c5361647d",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "_id",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "avh"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-318c5361647d",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\":\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "x"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-318c5361647d",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "exp",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "f75"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-318c5361647d",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "_",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Nini1"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-318c5361647d",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "004",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "MXB"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-318c5361647d",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\"}",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Vc4"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-318c5361647d",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": "tool_calls",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "rnph"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-318c5361647d",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": {
+            "completion_tokens": 19,
+            "prompt_tokens": 450,
+            "total_tokens": 469,
+            "completion_tokens_details": {
+              "accepted_prediction_tokens": 0,
+              "audio_tokens": 0,
+              "reasoning_tokens": 0,
+              "rejected_prediction_tokens": 0
+            },
+            "prompt_tokens_details": {
+              "audio_tokens": 0,
+              "cached_tokens": 0
+            }
+          },
+          "obfuscation": "nUptVmnQlQZrH"
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/recordings/42c357284497af596ae6c9341b0c189daa31e88b25d0381a985f24203b7a5a38.json

@@ -10,7 +10,7 @@
   },
   "response": {
     "body": {
-      "__type__": "llama_stack.apis.tools.tools.ToolInvocationResult",
+      "__type__": "llama_stack_api.tools.ToolInvocationResult",
       "__data__": {
         "content": "{\"query\": \"Llama 4 Maverick model experts\", \"top_k\": [{\"url\": \"https://console.groq.com/docs/model/meta-llama/llama-4-maverick-17b-128e-instruct\", \"title\": \"Llama 4 Maverick 17B 128E\", \"content\": \"Llama 4 Maverick is Meta's natively multimodal model that enables text and image understanding. With a 17 billion parameter mixture-of-experts architecture (128 experts), this model offers industry-leading performance for multimodal tasks like natural assistant-like chat, image recognition, and coding tasks. Llama 4 Maverick features an auto-regressive language model that uses a mixture-of-experts (MoE) architecture with 17B activated parameters (400B total) and incorporates early fusion for native multimodality. The model uses 128 experts to efficiently handle both text and image inputs while maintaining high performance across chat, knowledge, and code generation tasks, with a knowledge cutoff of August 2024. * For multimodal applications, this model supports up to 5 image inputs create(  model =\\\"meta-llama/llama-4-maverick-17b-128e-instruct\\\",   messages =[  {  \\\"role\\\":  \\\"user\\\",   \\\"content\\\":  \\\"Explain why fast inference is critical for reasoning models\\\"   }   ]  )  print(completion.\", \"score\": 0.9170729, \"raw_content\": null}, {\"url\": \"https://huggingface.co/meta-llama/Llama-4-Maverick-17B-128E\", \"title\": \"meta-llama/Llama-4-Maverick-17B-128E - Hugging Face\", \"content\": \"Model Architecture: The Llama 4 models are auto-regressive language models that use a mixture-of-experts (MoE) architecture and incorporate\", \"score\": 0.8021998, \"raw_content\": null}, {\"url\": \"https://www.ibm.com/new/announcements/meta-llama-4-maverick-and-llama-4-scout-now-available-in-watsonx-ai\", \"title\": \"Meta Llama 4 Maverick and Llama 4 Scout now available in watsonx ...\", \"content\": \"# Meta Llama 4 Maverick and Llama 4 Scout now available in watsonx.ai **IBM is excited to announce the addition of Meta\\u2019s latest generation of open models, Llama 4, to** **watsonx.ai****.** Llama 4 Scout and Llama 4 Maverick, the first mixture of experts (MoE) models released by Meta, provide frontier multimodal performance, high speeds, low cost, and industry leading context length. With the introduction of these latest offerings from Meta, IBM now supports a total of 13 Meta models in the expansive library of \\u00a0foundation models available in watsonx.ai. Trained on 40 trillion tokens of data, Llama 4 Scout offers performance rivalling or exceeding that of models with significantly larger active parameter counts while keeping costs and latency low. ## Llama 4 models on IBM watsonx\", \"score\": 0.78194773, \"raw_content\": null}, {\"url\": \"https://medium.com/@divyanshbhatiajm19/metas-llama-4-family-the-complete-guide-to-scout-maverick-and-behemoth-ai-models-in-2025-21a90c882e8a\", \"title\": \"Meta's Llama 4 Family: The Complete Guide to Scout, Maverick, and ...\", \"content\": \"# Meta\\u2019s Llama 4 Family: The Complete Guide to Scout, Maverick, and Behemoth AI Models in 2025 Feature Llama 4 Scout Llama 4 Maverick Llama 4 Behemoth **Total Parameters** 109B 400B ~2T **Active Parameters** 17B 17B 288B **Expert Count** 16 128 16 **Context Window** 10M tokens 1M tokens Not specified **Hardware Requirements** Single H100 GPU Single H100 DGX host Multiple GPUs **Inference Cost** Not specified $0.19-$0.49 per 1M tokens Not specified **Release Status** Available now Available now In training **Primary Use Cases** Long-context analysis, code processing High-performance multimodal applications Research, STEM reasoning The Llama 4 family represents Meta\\u2019s most significant AI development to date, with each model offering distinct advantages for different use cases:\", \"score\": 0.69672287, \"raw_content\": null}, {\"url\": \"https://www.llama.com/models/llama-4/\", \"title\": \"Unmatched Performance and Efficiency | Llama 4\", \"content\": \"# Llama 4 # Llama 4 Llama 4 Scout Class-leading natively multimodal model that offers superior text and visual intelligence, single H100 GPU efficiency, and a 10M context window for seamless long document analysis. Llama 4 MaverickIndustry-leading natively multimodal model for image and text understanding with groundbreaking intelligence and fast responses at a low cost. We evaluated model performance on a suite of common benchmarks across a wide range of languages, testing for coding, reasoning, knowledge, vision understanding, multilinguality, and long context. 4. Specialized long context evals are not traditionally reported for generalist models, so we share internal runs to showcase llama's frontier performance. 4. Specialized long context evals are not traditionally reported for generalist models, so we share internal runs to showcase llama's frontier performance.\", \"score\": 0.629889, \"raw_content\": null}]}",
         "error_message": null,

tests/integration/responses/recordings/430a49246c97c29bd958f383627f53ec795fd77ef818827e16691689151bf17c.json

@@ -0,0 +1,413 @@
+{
+  "test_id": "tests/integration/responses/test_tool_responses.py::test_response_non_streaming_multi_turn_tool_execution[openai_client-txt=openai/gpt-4o-user_file_access_check]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "I need to check if user 'alice' can access the file 'document.txt'. First, get alice's user ID, then check if that user ID can access the file 'document.txt'. Do this as a series of steps, where each step is a separate message. Return only one tool call per step. Summarize the final result with a single 'yes' or 'no' response."
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      },
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "get_user_id",
+            "description": "\n        Get the user ID for a given username. This ID is needed for other operations.\n\n        :param username: The username to look up\n        :return: The user ID for the username\n        ",
+            "parameters": {
+              "properties": {
+                "username": {
+                  "title": "Username",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "username"
+              ],
+              "title": "get_user_idArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_user_permissions",
+            "description": "\n        Get the permissions for a user ID. Requires a valid user ID from get_user_id.\n\n        :param user_id: The user ID to check permissions for\n        :return: The permissions for the user\n        ",
+            "parameters": {
+              "properties": {
+                "user_id": {
+                  "title": "User Id",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "user_id"
+              ],
+              "title": "get_user_permissionsArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "check_file_access",
+            "description": "\n        Check if a user can access a specific file. Requires a valid user ID.\n\n        :param user_id: The user ID to check access for\n        :param filename: The filename to check access to\n        :return: Whether the user can access the file (yes/no)\n        ",
+            "parameters": {
+              "properties": {
+                "user_id": {
+                  "title": "User Id",
+                  "type": "string"
+                },
+                "filename": {
+                  "title": "Filename",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "user_id",
+                "filename"
+              ],
+              "title": "check_file_accessArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_experiment_id",
+            "description": "\n        Get the experiment ID for a given experiment name. This ID is needed to get results.\n\n        :param experiment_name: The name of the experiment\n        :return: The experiment ID\n        ",
+            "parameters": {
+              "properties": {
+                "experiment_name": {
+                  "title": "Experiment Name",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "experiment_name"
+              ],
+              "title": "get_experiment_idArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_experiment_results",
+            "description": "\n        Get the results for an experiment ID. Requires a valid experiment ID from get_experiment_id.\n\n        :param experiment_id: The experiment ID to get results for\n        :return: The experiment results\n        ",
+            "parameters": {
+              "properties": {
+                "experiment_id": {
+                  "title": "Experiment Id",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "experiment_id"
+              ],
+              "title": "get_experiment_resultsArguments",
+              "type": "object"
+            }
+          }
+        }
+      ]
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-430a49246c97",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": "call_EsVvmBUqtJb42kNkYnK19QkJ",
+                    "function": {
+                      "arguments": "",
+                      "name": "get_user_id"
+                    },
+                    "type": "function"
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Ma7aiZxSs"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-430a49246c97",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "{\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "DXu"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-430a49246c97",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "username",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "rtfrl7gxu80vmN"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-430a49246c97",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\":\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "r"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-430a49246c97",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "alice",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "M"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-430a49246c97",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\"}",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "vSu"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-430a49246c97",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": "tool_calls",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "sXfh"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-430a49246c97",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": {
+            "completion_tokens": 15,
+            "prompt_tokens": 454,
+            "total_tokens": 469,
+            "completion_tokens_details": {
+              "accepted_prediction_tokens": 0,
+              "audio_tokens": 0,
+              "reasoning_tokens": 0,
+              "rejected_prediction_tokens": 0
+            },
+            "prompt_tokens_details": {
+              "audio_tokens": 0,
+              "cached_tokens": 0
+            }
+          },
+          "obfuscation": "bEe7hWJ6U62YQ"
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/recordings/51e3ddbc9d23c614ead9a8fd6ad30294237eb43063c00efc83b8a1202c1cc20c.json

@@ -0,0 +1,614 @@
+{
+  "test_id": "tests/integration/responses/test_mcp_authentication.py::test_mcp_authorization_backward_compatibility[openai_client-txt=openai/gpt-4o]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "What is the boiling point of myawesomeliquid?"
+        },
+        {
+          "role": "assistant",
+          "content": "",
+          "tool_calls": [
+            {
+              "index": 0,
+              "id": "call_UeAsx9M8mAXo1F1LZj6TsEV9",
+              "type": "function",
+              "function": {
+                "name": "get_boiling_point",
+                "arguments": "{\"liquid_name\":\"myawesomeliquid\"}"
+              }
+            }
+          ]
+        },
+        {
+          "role": "tool",
+          "tool_call_id": "call_UeAsx9M8mAXo1F1LZj6TsEV9",
+          "content": [
+            {
+              "type": "text",
+              "text": "-100"
+            }
+          ]
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      },
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "greet_everyone",
+            "parameters": {
+              "properties": {
+                "url": {
+                  "title": "Url",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "url"
+              ],
+              "title": "greet_everyoneArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_boiling_point",
+            "description": "\n        Returns the boiling point of a liquid in Celsius or Fahrenheit.\n\n        :param liquid_name: The name of the liquid\n        :param celsius: Whether to return the boiling point in Celsius\n        :return: The boiling point of the liquid in Celcius or Fahrenheit\n        ",
+            "parameters": {
+              "properties": {
+                "liquid_name": {
+                  "title": "Liquid Name",
+                  "type": "string"
+                },
+                "celsius": {
+                  "default": true,
+                  "title": "Celsius",
+                  "type": "boolean"
+                }
+              },
+              "required": [
+                "liquid_name"
+              ],
+              "title": "get_boiling_pointArguments",
+              "type": "object"
+            }
+          }
+        }
+      ]
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-51e3ddbc9d23",
+          "choices": [
+            {
+              "delta": {
+                "content": "",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "c5g42LQpiBwmVH"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-51e3ddbc9d23",
+          "choices": [
+            {
+              "delta": {
+                "content": "The",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "MEmQFjCKEsNDL"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-51e3ddbc9d23",
+          "choices": [
+            {
+              "delta": {
+                "content": " boiling",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "dF3UemYO"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-51e3ddbc9d23",
+          "choices": [
+            {
+              "delta": {
+                "content": " point",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "ENDOmjG37D"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-51e3ddbc9d23",
+          "choices": [
+            {
+              "delta": {
+                "content": " of",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "6kb5u2d4ILV59"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-51e3ddbc9d23",
+          "choices": [
+            {
+              "delta": {
+                "content": " \"",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Y6Dp6rbT9OdBG"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-51e3ddbc9d23",
+          "choices": [
+            {
+              "delta": {
+                "content": "my",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "EN0ShAkdxF2jIs"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-51e3ddbc9d23",
+          "choices": [
+            {
+              "delta": {
+                "content": "aw",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "1NHavCOT2fSI63"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-51e3ddbc9d23",
+          "choices": [
+            {
+              "delta": {
+                "content": "esom",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "VTwbnRFtKY2W"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-51e3ddbc9d23",
+          "choices": [
+            {
+              "delta": {
+                "content": "eli",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "VJuNhLeGK43e6"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-51e3ddbc9d23",
+          "choices": [
+            {
+              "delta": {
+                "content": "quid",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "bFgxcYCjU42I"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-51e3ddbc9d23",
+          "choices": [
+            {
+              "delta": {
+                "content": "\"",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "5KR4mGTP0Rpu0O"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-51e3ddbc9d23",
+          "choices": [
+            {
+              "delta": {
+                "content": " is",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "KCeY3i4Qo9L1j"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-51e3ddbc9d23",
+          "choices": [
+            {
+              "delta": {
+                "content": " -",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "GgtT2kqCUk8jGH"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-51e3ddbc9d23",
+          "choices": [
+            {
+              "delta": {
+                "content": "100",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "H3E18AkuuATh3"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-51e3ddbc9d23",
+          "choices": [
+            {
+              "delta": {
+                "content": "\u00b0C",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "5kuUoomGw6aPf0"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-51e3ddbc9d23",
+          "choices": [
+            {
+              "delta": {
+                "content": ".",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "CKIiDxWMV3zzcNj"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-51e3ddbc9d23",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": "stop",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "9KZoS4rawE"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-51e3ddbc9d23",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": {
+            "completion_tokens": 17,
+            "prompt_tokens": 188,
+            "total_tokens": 205,
+            "completion_tokens_details": {
+              "accepted_prediction_tokens": 0,
+              "audio_tokens": 0,
+              "reasoning_tokens": 0,
+              "rejected_prediction_tokens": 0
+            },
+            "prompt_tokens_details": {
+              "audio_tokens": 0,
+              "cached_tokens": 0
+            }
+          },
+          "obfuscation": "iq2ecCxqopvPO"
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/recordings/5236eb1d546e5a1bd0712891d8b4866a73cc04ce93db40346beb070f30fafee1.json

@@ -0,0 +1,614 @@
+{
+  "test_id": "tests/integration/responses/test_mcp_authentication.py::test_mcp_authorization_bearer[client_with_models-txt=openai/gpt-4o]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "What is the boiling point of myawesomeliquid?"
+        },
+        {
+          "role": "assistant",
+          "content": "",
+          "tool_calls": [
+            {
+              "index": 0,
+              "id": "call_mitVYvmPaFfoSmKjzKo5xmZp",
+              "type": "function",
+              "function": {
+                "name": "get_boiling_point",
+                "arguments": "{\"liquid_name\":\"myawesomeliquid\"}"
+              }
+            }
+          ]
+        },
+        {
+          "role": "tool",
+          "tool_call_id": "call_mitVYvmPaFfoSmKjzKo5xmZp",
+          "content": [
+            {
+              "type": "text",
+              "text": "-100"
+            }
+          ]
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      },
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "greet_everyone",
+            "parameters": {
+              "properties": {
+                "url": {
+                  "title": "Url",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "url"
+              ],
+              "title": "greet_everyoneArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_boiling_point",
+            "description": "\n        Returns the boiling point of a liquid in Celsius or Fahrenheit.\n\n        :param liquid_name: The name of the liquid\n        :param celsius: Whether to return the boiling point in Celsius\n        :return: The boiling point of the liquid in Celcius or Fahrenheit\n        ",
+            "parameters": {
+              "properties": {
+                "liquid_name": {
+                  "title": "Liquid Name",
+                  "type": "string"
+                },
+                "celsius": {
+                  "default": true,
+                  "title": "Celsius",
+                  "type": "boolean"
+                }
+              },
+              "required": [
+                "liquid_name"
+              ],
+              "title": "get_boiling_pointArguments",
+              "type": "object"
+            }
+          }
+        }
+      ]
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5236eb1d546e",
+          "choices": [
+            {
+              "delta": {
+                "content": "",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "veiGKPHTdRNcOX"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5236eb1d546e",
+          "choices": [
+            {
+              "delta": {
+                "content": "The",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "u9RK8eZYDguJs"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5236eb1d546e",
+          "choices": [
+            {
+              "delta": {
+                "content": " boiling",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "U0L1RjHF"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5236eb1d546e",
+          "choices": [
+            {
+              "delta": {
+                "content": " point",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "TMS6QVLJfj"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5236eb1d546e",
+          "choices": [
+            {
+              "delta": {
+                "content": " of",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "5zokjwZ0nBNlD"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5236eb1d546e",
+          "choices": [
+            {
+              "delta": {
+                "content": " \"",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "CmOp3DQRu0AqZ"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5236eb1d546e",
+          "choices": [
+            {
+              "delta": {
+                "content": "my",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "OlnZU0jlGyE2mD"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5236eb1d546e",
+          "choices": [
+            {
+              "delta": {
+                "content": "aw",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "PGCsCfw8zUqRAj"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5236eb1d546e",
+          "choices": [
+            {
+              "delta": {
+                "content": "esom",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "8P65fJ4x3QVF"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5236eb1d546e",
+          "choices": [
+            {
+              "delta": {
+                "content": "eli",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "HVTNGb62o54Ol"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5236eb1d546e",
+          "choices": [
+            {
+              "delta": {
+                "content": "quid",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "bdRgQioKQZM6"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5236eb1d546e",
+          "choices": [
+            {
+              "delta": {
+                "content": "\"",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "5djjyePEzwsPID"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5236eb1d546e",
+          "choices": [
+            {
+              "delta": {
+                "content": " is",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "xoN3TaCEum6A9"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5236eb1d546e",
+          "choices": [
+            {
+              "delta": {
+                "content": " -",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "UmU8LCL6WJIDrf"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5236eb1d546e",
+          "choices": [
+            {
+              "delta": {
+                "content": "100",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "FFXxvyme7JKyc"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5236eb1d546e",
+          "choices": [
+            {
+              "delta": {
+                "content": "\u00b0C",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "8BpDPmgFmIBJQQ"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5236eb1d546e",
+          "choices": [
+            {
+              "delta": {
+                "content": ".",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Mey7rwshfBQbVlP"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5236eb1d546e",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": "stop",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "IXaz4vn8As"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5236eb1d546e",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": {
+            "completion_tokens": 17,
+            "prompt_tokens": 188,
+            "total_tokens": 205,
+            "completion_tokens_details": {
+              "accepted_prediction_tokens": 0,
+              "audio_tokens": 0,
+              "reasoning_tokens": 0,
+              "rejected_prediction_tokens": 0
+            },
+            "prompt_tokens_details": {
+              "audio_tokens": 0,
+              "cached_tokens": 0
+            }
+          },
+          "obfuscation": "9ebnd6bFXcdOY"
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/recordings/52a2b96781961e252aa3a7b0a5ff77eb5d0989d312e929ed59dda07738487d09.json

@@ -0,0 +1,586 @@
+{
+  "test_id": "tests/integration/responses/test_tool_responses.py::test_response_streaming_multi_turn_tool_execution[openai_client-txt=openai/gpt-4o-user_permissions_workflow]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "Help me with this security check: First, get the user ID for 'charlie', then get the permissions for that user ID, and finally check if that user can access 'secret_file.txt'. Stream your progress as you work through each step. Return only one tool call per step. Summarize the final result with a single 'yes' or 'no' response."
+        },
+        {
+          "role": "assistant",
+          "content": "",
+          "tool_calls": [
+            {
+              "index": 0,
+              "id": "call_fsxGbKmceUbLSXCe4sx9WLXO",
+              "type": "function",
+              "function": {
+                "name": "get_user_id",
+                "arguments": "{\"username\":\"charlie\"}"
+              }
+            }
+          ]
+        },
+        {
+          "role": "tool",
+          "tool_call_id": "call_fsxGbKmceUbLSXCe4sx9WLXO",
+          "content": [
+            {
+              "type": "text",
+              "text": "user_11111"
+            }
+          ]
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      },
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "get_user_id",
+            "description": "\n        Get the user ID for a given username. This ID is needed for other operations.\n\n        :param username: The username to look up\n        :return: The user ID for the username\n        ",
+            "parameters": {
+              "properties": {
+                "username": {
+                  "title": "Username",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "username"
+              ],
+              "title": "get_user_idArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_user_permissions",
+            "description": "\n        Get the permissions for a user ID. Requires a valid user ID from get_user_id.\n\n        :param user_id: The user ID to check permissions for\n        :return: The permissions for the user\n        ",
+            "parameters": {
+              "properties": {
+                "user_id": {
+                  "title": "User Id",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "user_id"
+              ],
+              "title": "get_user_permissionsArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "check_file_access",
+            "description": "\n        Check if a user can access a specific file. Requires a valid user ID.\n\n        :param user_id: The user ID to check access for\n        :param filename: The filename to check access to\n        :return: Whether the user can access the file (yes/no)\n        ",
+            "parameters": {
+              "properties": {
+                "user_id": {
+                  "title": "User Id",
+                  "type": "string"
+                },
+                "filename": {
+                  "title": "Filename",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "user_id",
+                "filename"
+              ],
+              "title": "check_file_accessArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_experiment_id",
+            "description": "\n        Get the experiment ID for a given experiment name. This ID is needed to get results.\n\n        :param experiment_name: The name of the experiment\n        :return: The experiment ID\n        ",
+            "parameters": {
+              "properties": {
+                "experiment_name": {
+                  "title": "Experiment Name",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "experiment_name"
+              ],
+              "title": "get_experiment_idArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_experiment_results",
+            "description": "\n        Get the results for an experiment ID. Requires a valid experiment ID from get_experiment_id.\n\n        :param experiment_id: The experiment ID to get results for\n        :return: The experiment results\n        ",
+            "parameters": {
+              "properties": {
+                "experiment_id": {
+                  "title": "Experiment Id",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "experiment_id"
+              ],
+              "title": "get_experiment_resultsArguments",
+              "type": "object"
+            }
+          }
+        }
+      ]
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-52a2b9678196",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": "call_moRBxqnBJ48EWTSEoQ1llgib",
+                    "function": {
+                      "arguments": "",
+                      "name": "get_user_permissions"
+                    },
+                    "type": "function"
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": ""
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-52a2b9678196",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "{\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "00p"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-52a2b9678196",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "user",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Y0"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-52a2b9678196",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "_id",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "i2I"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-52a2b9678196",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\":\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "P"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-52a2b9678196",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "user",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "IG"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-52a2b9678196",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "_",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "QY61l"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-52a2b9678196",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "111",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "YAZ"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-52a2b9678196",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "11",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Nw7U"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-52a2b9678196",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\"}",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Ev7"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-52a2b9678196",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": "tool_calls",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "CSaD"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-52a2b9678196",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": {
+            "completion_tokens": 19,
+            "prompt_tokens": 478,
+            "total_tokens": 497,
+            "completion_tokens_details": {
+              "accepted_prediction_tokens": 0,
+              "audio_tokens": 0,
+              "reasoning_tokens": 0,
+              "rejected_prediction_tokens": 0
+            },
+            "prompt_tokens_details": {
+              "audio_tokens": 0,
+              "cached_tokens": 0
+            }
+          },
+          "obfuscation": "kMNEyeKFT75vK"
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/recordings/541b5db7789e61d2400b70bd41c2ff7145784d249c3216c34299c38c28118328.json

@@ -0,0 +1,524 @@
+{
+  "test_id": "tests/integration/responses/test_tool_responses.py::test_response_non_streaming_multi_turn_tool_execution[openai_client-txt=openai/gpt-4o-experiment_results_lookup]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "I need to get the results for the 'boiling_point' experiment. First, get the experiment ID for 'boiling_point', then use that ID to get the experiment results. Tell me the boiling point in Celsius."
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      },
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "get_user_id",
+            "description": "\n        Get the user ID for a given username. This ID is needed for other operations.\n\n        :param username: The username to look up\n        :return: The user ID for the username\n        ",
+            "parameters": {
+              "properties": {
+                "username": {
+                  "title": "Username",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "username"
+              ],
+              "title": "get_user_idArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_user_permissions",
+            "description": "\n        Get the permissions for a user ID. Requires a valid user ID from get_user_id.\n\n        :param user_id: The user ID to check permissions for\n        :return: The permissions for the user\n        ",
+            "parameters": {
+              "properties": {
+                "user_id": {
+                  "title": "User Id",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "user_id"
+              ],
+              "title": "get_user_permissionsArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "check_file_access",
+            "description": "\n        Check if a user can access a specific file. Requires a valid user ID.\n\n        :param user_id: The user ID to check access for\n        :param filename: The filename to check access to\n        :return: Whether the user can access the file (yes/no)\n        ",
+            "parameters": {
+              "properties": {
+                "user_id": {
+                  "title": "User Id",
+                  "type": "string"
+                },
+                "filename": {
+                  "title": "Filename",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "user_id",
+                "filename"
+              ],
+              "title": "check_file_accessArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_experiment_id",
+            "description": "\n        Get the experiment ID for a given experiment name. This ID is needed to get results.\n\n        :param experiment_name: The name of the experiment\n        :return: The experiment ID\n        ",
+            "parameters": {
+              "properties": {
+                "experiment_name": {
+                  "title": "Experiment Name",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "experiment_name"
+              ],
+              "title": "get_experiment_idArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_experiment_results",
+            "description": "\n        Get the results for an experiment ID. Requires a valid experiment ID from get_experiment_id.\n\n        :param experiment_id: The experiment ID to get results for\n        :return: The experiment results\n        ",
+            "parameters": {
+              "properties": {
+                "experiment_id": {
+                  "title": "Experiment Id",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "experiment_id"
+              ],
+              "title": "get_experiment_resultsArguments",
+              "type": "object"
+            }
+          }
+        }
+      ]
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-541b5db7789e",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": "call_dZwjBxH3aTRhnaS0bJVPqRcz",
+                    "function": {
+                      "arguments": "",
+                      "name": "get_experiment_id"
+                    },
+                    "type": "function"
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "W3B"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-541b5db7789e",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "{\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "L7n"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-541b5db7789e",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "experiment",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "lXUc0FKJkRea"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-541b5db7789e",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "_name",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "0"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-541b5db7789e",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\":\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "D"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-541b5db7789e",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "bo",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "3dUQ"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-541b5db7789e",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "iling",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "1"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-541b5db7789e",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "_point",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": ""
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-541b5db7789e",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\"}",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "48i"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-541b5db7789e",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": "tool_calls",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "eQyU"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-541b5db7789e",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": {
+            "completion_tokens": 19,
+            "prompt_tokens": 418,
+            "total_tokens": 437,
+            "completion_tokens_details": {
+              "accepted_prediction_tokens": 0,
+              "audio_tokens": 0,
+              "reasoning_tokens": 0,
+              "rejected_prediction_tokens": 0
+            },
+            "prompt_tokens_details": {
+              "audio_tokens": 0,
+              "cached_tokens": 0
+            }
+          },
+          "obfuscation": "5tVrc5IEigum8"
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/recordings/54aa690e31b5c33a0488a5d7403393e5712917253462292829b37b9320d6df82.json

@@ -10,7 +10,7 @@
   },
   "response": {
     "body": {
-      "__type__": "llama_stack.apis.tools.tools.ToolInvocationResult",
+      "__type__": "llama_stack_api.tools.ToolInvocationResult",
       "__data__": {
         "content": "{\"query\": \"Llama 4 Maverick model number of experts\", \"top_k\": [{\"url\": \"https://console.groq.com/docs/model/meta-llama/llama-4-maverick-17b-128e-instruct\", \"title\": \"Llama 4 Maverick 17B 128E\", \"content\": \"Llama 4 Maverick is Meta's natively multimodal model that enables text and image understanding. With a 17 billion parameter mixture-of-experts architecture (128 experts), this model offers industry-leading performance for multimodal tasks like natural assistant-like chat, image recognition, and coding tasks. Llama 4 Maverick features an auto-regressive language model that uses a mixture-of-experts (MoE) architecture with 17B activated parameters (400B total) and incorporates early fusion for native multimodality. The model uses 128 experts to efficiently handle both text and image inputs while maintaining high performance across chat, knowledge, and code generation tasks, with a knowledge cutoff of August 2024. * For multimodal applications, this model supports up to 5 image inputs create(  model =\\\"meta-llama/llama-4-maverick-17b-128e-instruct\\\",   messages =[  {  \\\"role\\\":  \\\"user\\\",   \\\"content\\\":  \\\"Explain why fast inference is critical for reasoning models\\\"   }   ]  )  print(completion.\", \"score\": 0.9287263, \"raw_content\": null}, {\"url\": \"https://huggingface.co/meta-llama/Llama-4-Maverick-17B-128E\", \"title\": \"meta-llama/Llama-4-Maverick-17B-128E\", \"content\": \"... model with 16 experts, and Llama 4 Maverick, a 17 billion parameter model with 128 experts. Model developer: Meta. Model Architecture: The\", \"score\": 0.9183121, \"raw_content\": null}, {\"url\": \"https://build.nvidia.com/meta/llama-4-maverick-17b-128e-instruct/modelcard\", \"title\": \"llama-4-maverick-17b-128e-instruct Model by Meta\", \"content\": \"... model with 16 experts, and Llama 4 Maverick, a 17 billion parameter model with 128 experts. Third-Party Community Consideration. This model\", \"score\": 0.91399205, \"raw_content\": null}, {\"url\": \"https://replicate.com/meta/llama-4-maverick-instruct\", \"title\": \"meta/llama-4-maverick-instruct | Run with an API on ...\", \"content\": \"... model with 16 experts, and Llama 4 Maverick, a 17 billion parameter model with 128 experts. All services are online \\u00b7 Home \\u00b7 About \\u00b7 Changelog\", \"score\": 0.9073207, \"raw_content\": null}, {\"url\": \"https://openrouter.ai/meta-llama/llama-4-maverick\", \"title\": \"Llama 4 Maverick - API, Providers, Stats\", \"content\": \"# Meta: Llama 4 Maverick ### meta-llama/llama-4-maverick Llama 4 Maverick 17B Instruct (128E) is a high-capacity multimodal language model from Meta, built on a mixture-of-experts (MoE) architecture with 128 experts and 17 billion active parameters per forward pass (400B total). Released on April 5, 2025 under the Llama 4 Community License, Maverick is suited for research and commercial applications requiring advanced multimodal understanding and high model throughput. Llama 4 Maverick - API, Providers, Stats | OpenRouter ## Providers for Llama 4 Maverick ## Performance for Llama 4 Maverick ## Apps using Llama 4 Maverick ## Recent activity on Llama 4 Maverick ## Uptime stats for Llama 4 Maverick ## Sample code and API for Llama 4 Maverick\", \"score\": 0.8958969, \"raw_content\": null}]}",
         "error_message": null,

tests/integration/responses/recordings/56ddb450d81590f461113ec5a55d0532e8f5b9418b22e5f874afff695601da16.json

@@ -0,0 +1,574 @@
+{
+  "test_id": "tests/integration/responses/test_mcp_authentication.py::test_mcp_authorization_backward_compatibility[openai_client-txt=openai/gpt-4o]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "What is the boiling point of myawesomeliquid?"
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      },
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "greet_everyone",
+            "parameters": {
+              "properties": {
+                "url": {
+                  "title": "Url",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "url"
+              ],
+              "title": "greet_everyoneArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_boiling_point",
+            "description": "\n        Returns the boiling point of a liquid in Celsius or Fahrenheit.\n\n        :param liquid_name: The name of the liquid\n        :param celsius: Whether to return the boiling point in Celsius\n        :return: The boiling point of the liquid in Celcius or Fahrenheit\n        ",
+            "parameters": {
+              "properties": {
+                "liquid_name": {
+                  "title": "Liquid Name",
+                  "type": "string"
+                },
+                "celsius": {
+                  "default": true,
+                  "title": "Celsius",
+                  "type": "boolean"
+                }
+              },
+              "required": [
+                "liquid_name"
+              ],
+              "title": "get_boiling_pointArguments",
+              "type": "object"
+            }
+          }
+        }
+      ]
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-56ddb450d815",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": "call_UeAsx9M8mAXo1F1LZj6TsEV9",
+                    "function": {
+                      "arguments": "",
+                      "name": "get_boiling_point"
+                    },
+                    "type": "function"
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "bKe"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-56ddb450d815",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "{\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "kxw"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-56ddb450d815",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "li",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "cKkF"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-56ddb450d815",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "quid",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "md"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-56ddb450d815",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "_name",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "O"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-56ddb450d815",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\":\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "o"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-56ddb450d815",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "my",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "nRfv"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-56ddb450d815",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "aw",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "1M8i"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-56ddb450d815",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "esom",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "7q"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-56ddb450d815",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "eli",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "R2Q"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-56ddb450d815",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "quid",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "lB"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-56ddb450d815",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\"}",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "MDi"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-56ddb450d815",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": "tool_calls",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "7KwE"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-56ddb450d815",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": {
+            "completion_tokens": 22,
+            "prompt_tokens": 154,
+            "total_tokens": 176,
+            "completion_tokens_details": {
+              "accepted_prediction_tokens": 0,
+              "audio_tokens": 0,
+              "reasoning_tokens": 0,
+              "rejected_prediction_tokens": 0
+            },
+            "prompt_tokens_details": {
+              "audio_tokens": 0,
+              "cached_tokens": 0
+            }
+          },
+          "obfuscation": "9IipvPESur5Y7"
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/recordings/59faeeca84b137e9b2c7d310ea47dc01025aeb2ee6203ef478133313e0a0e250.json

@@ -0,0 +1,614 @@
+{
+  "test_id": "tests/integration/responses/test_mcp_authentication.py::test_mcp_authorization_bearer[openai_client-txt=openai/gpt-4o]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "What is the boiling point of myawesomeliquid?"
+        },
+        {
+          "role": "assistant",
+          "content": "",
+          "tool_calls": [
+            {
+              "index": 0,
+              "id": "call_2lYntxgdJV66JFvD6OuICQCB",
+              "type": "function",
+              "function": {
+                "name": "get_boiling_point",
+                "arguments": "{\"liquid_name\":\"myawesomeliquid\"}"
+              }
+            }
+          ]
+        },
+        {
+          "role": "tool",
+          "tool_call_id": "call_2lYntxgdJV66JFvD6OuICQCB",
+          "content": [
+            {
+              "type": "text",
+              "text": "-100"
+            }
+          ]
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      },
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "greet_everyone",
+            "parameters": {
+              "properties": {
+                "url": {
+                  "title": "Url",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "url"
+              ],
+              "title": "greet_everyoneArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_boiling_point",
+            "description": "\n        Returns the boiling point of a liquid in Celsius or Fahrenheit.\n\n        :param liquid_name: The name of the liquid\n        :param celsius: Whether to return the boiling point in Celsius\n        :return: The boiling point of the liquid in Celcius or Fahrenheit\n        ",
+            "parameters": {
+              "properties": {
+                "liquid_name": {
+                  "title": "Liquid Name",
+                  "type": "string"
+                },
+                "celsius": {
+                  "default": true,
+                  "title": "Celsius",
+                  "type": "boolean"
+                }
+              },
+              "required": [
+                "liquid_name"
+              ],
+              "title": "get_boiling_pointArguments",
+              "type": "object"
+            }
+          }
+        }
+      ]
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-59faeeca84b1",
+          "choices": [
+            {
+              "delta": {
+                "content": "",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "BNpFmbWkpYEjZX"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-59faeeca84b1",
+          "choices": [
+            {
+              "delta": {
+                "content": "The",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "HdnyHcq2CLvjn"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-59faeeca84b1",
+          "choices": [
+            {
+              "delta": {
+                "content": " boiling",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "gOMuwgrp"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-59faeeca84b1",
+          "choices": [
+            {
+              "delta": {
+                "content": " point",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "OTfqq7Yggw"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-59faeeca84b1",
+          "choices": [
+            {
+              "delta": {
+                "content": " of",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "cwJMhZJyf5PIp"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-59faeeca84b1",
+          "choices": [
+            {
+              "delta": {
+                "content": " \"",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "54NR7IGiuBTw5"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-59faeeca84b1",
+          "choices": [
+            {
+              "delta": {
+                "content": "my",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "q1x9cVVPTflQti"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-59faeeca84b1",
+          "choices": [
+            {
+              "delta": {
+                "content": "aw",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "vcudLe3yaadkvB"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-59faeeca84b1",
+          "choices": [
+            {
+              "delta": {
+                "content": "esom",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "uql1pBt4elRL"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-59faeeca84b1",
+          "choices": [
+            {
+              "delta": {
+                "content": "eli",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "M2kzUEkJctjYp"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-59faeeca84b1",
+          "choices": [
+            {
+              "delta": {
+                "content": "quid",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Waet2ux2zs9P"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-59faeeca84b1",
+          "choices": [
+            {
+              "delta": {
+                "content": "\"",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "KjbjxdGYUZDuiI"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-59faeeca84b1",
+          "choices": [
+            {
+              "delta": {
+                "content": " is",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Fg8IXJhJv8iAI"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-59faeeca84b1",
+          "choices": [
+            {
+              "delta": {
+                "content": " -",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "wiAqPLAoinVhQq"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-59faeeca84b1",
+          "choices": [
+            {
+              "delta": {
+                "content": "100",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "vJnb9sE969jph"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-59faeeca84b1",
+          "choices": [
+            {
+              "delta": {
+                "content": "\u00b0C",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "5Hgi5CU0aV0sPw"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-59faeeca84b1",
+          "choices": [
+            {
+              "delta": {
+                "content": ".",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "RDfKhuQo4E4TLXU"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-59faeeca84b1",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": "stop",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "oN1EYVkDbW"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-59faeeca84b1",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": {
+            "completion_tokens": 17,
+            "prompt_tokens": 188,
+            "total_tokens": 205,
+            "completion_tokens_details": {
+              "accepted_prediction_tokens": 0,
+              "audio_tokens": 0,
+              "reasoning_tokens": 0,
+              "rejected_prediction_tokens": 0
+            },
+            "prompt_tokens_details": {
+              "audio_tokens": 0,
+              "cached_tokens": 0
+            }
+          },
+          "obfuscation": "OfhOTT3VdJ2s7"
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/recordings/6a05cad89f138e215047fd44d21803c4a397f772ad8b1cb90ec44527ce964a45.json

@@ -0,0 +1,614 @@
+{
+  "test_id": "tests/integration/responses/test_tool_responses.py::test_response_non_streaming_mcp_tool[openai_client-txt=openai/gpt-4o-boiling_point_tool]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "What is the boiling point of myawesomeliquid in Celsius?"
+        },
+        {
+          "role": "assistant",
+          "content": "",
+          "tool_calls": [
+            {
+              "index": 0,
+              "id": "call_8kf8fNIDcWOelbCmUEcretON",
+              "type": "function",
+              "function": {
+                "name": "get_boiling_point",
+                "arguments": "{\"liquid_name\":\"myawesomeliquid\",\"celsius\":true}"
+              }
+            }
+          ]
+        },
+        {
+          "role": "tool",
+          "tool_call_id": "call_8kf8fNIDcWOelbCmUEcretON",
+          "content": [
+            {
+              "type": "text",
+              "text": "-100"
+            }
+          ]
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      },
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "greet_everyone",
+            "parameters": {
+              "properties": {
+                "url": {
+                  "title": "Url",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "url"
+              ],
+              "title": "greet_everyoneArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_boiling_point",
+            "description": "\n        Returns the boiling point of a liquid in Celsius or Fahrenheit.\n\n        :param liquid_name: The name of the liquid\n        :param celsius: Whether to return the boiling point in Celsius\n        :return: The boiling point of the liquid in Celcius or Fahrenheit\n        ",
+            "parameters": {
+              "properties": {
+                "liquid_name": {
+                  "title": "Liquid Name",
+                  "type": "string"
+                },
+                "celsius": {
+                  "default": true,
+                  "title": "Celsius",
+                  "type": "boolean"
+                }
+              },
+              "required": [
+                "liquid_name"
+              ],
+              "title": "get_boiling_pointArguments",
+              "type": "object"
+            }
+          }
+        }
+      ]
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-6a05cad89f13",
+          "choices": [
+            {
+              "delta": {
+                "content": "",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "QvigjcdULEdran"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-6a05cad89f13",
+          "choices": [
+            {
+              "delta": {
+                "content": "The",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "sIHyVud88f1Ri"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-6a05cad89f13",
+          "choices": [
+            {
+              "delta": {
+                "content": " boiling",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "L46IcJeM"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-6a05cad89f13",
+          "choices": [
+            {
+              "delta": {
+                "content": " point",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "j0afpRCRBL"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-6a05cad89f13",
+          "choices": [
+            {
+              "delta": {
+                "content": " of",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "tuzBzZB7jURPj"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-6a05cad89f13",
+          "choices": [
+            {
+              "delta": {
+                "content": " \"",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "iq6vUNVBRuRH5"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-6a05cad89f13",
+          "choices": [
+            {
+              "delta": {
+                "content": "my",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Nkkz9uUPfhHdqZ"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-6a05cad89f13",
+          "choices": [
+            {
+              "delta": {
+                "content": "aw",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "oR3PEQpsXLwYOJ"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-6a05cad89f13",
+          "choices": [
+            {
+              "delta": {
+                "content": "esom",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "VBFf1ewix1rj"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-6a05cad89f13",
+          "choices": [
+            {
+              "delta": {
+                "content": "eli",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "yEx3rYoaZjsTw"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-6a05cad89f13",
+          "choices": [
+            {
+              "delta": {
+                "content": "quid",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "I6VR8wzPmnpa"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-6a05cad89f13",
+          "choices": [
+            {
+              "delta": {
+                "content": "\"",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "xld69F07KIb2Yc"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-6a05cad89f13",
+          "choices": [
+            {
+              "delta": {
+                "content": " is",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "GKgtQZJiWLVKj"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-6a05cad89f13",
+          "choices": [
+            {
+              "delta": {
+                "content": " -",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "1by4tgiJqNgaI1"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-6a05cad89f13",
+          "choices": [
+            {
+              "delta": {
+                "content": "100",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "2RdP6HDQApUpN"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-6a05cad89f13",
+          "choices": [
+            {
+              "delta": {
+                "content": "\u00b0C",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "21ABialEpJBCcX"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-6a05cad89f13",
+          "choices": [
+            {
+              "delta": {
+                "content": ".",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "uoaaRgmiGLD815k"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-6a05cad89f13",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": "stop",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "QKEKTjUUam"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-6a05cad89f13",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": {
+            "completion_tokens": 17,
+            "prompt_tokens": 195,
+            "total_tokens": 212,
+            "completion_tokens_details": {
+              "accepted_prediction_tokens": 0,
+              "audio_tokens": 0,
+              "reasoning_tokens": 0,
+              "rejected_prediction_tokens": 0
+            },
+            "prompt_tokens_details": {
+              "audio_tokens": 0,
+              "cached_tokens": 0
+            }
+          },
+          "obfuscation": "ceWQr6uzZRuj3"
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/recordings/6d7f54b7be4845c31ae64498e8018a218bb7f4b8363998abc34ec9bb7ba3a03d.json

@@ -0,0 +1,574 @@
+{
+  "test_id": "tests/integration/responses/test_tool_responses.py::test_response_mcp_tool_approval[openai_client-txt=openai/gpt-4o-False-boiling_point_tool]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "What is the boiling point of myawesomeliquid in Celsius?"
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      },
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "greet_everyone",
+            "parameters": {
+              "properties": {
+                "url": {
+                  "title": "Url",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "url"
+              ],
+              "title": "greet_everyoneArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_boiling_point",
+            "description": "\n        Returns the boiling point of a liquid in Celsius or Fahrenheit.\n\n        :param liquid_name: The name of the liquid\n        :param celsius: Whether to return the boiling point in Celsius\n        :return: The boiling point of the liquid in Celcius or Fahrenheit\n        ",
+            "parameters": {
+              "properties": {
+                "liquid_name": {
+                  "title": "Liquid Name",
+                  "type": "string"
+                },
+                "celsius": {
+                  "default": true,
+                  "title": "Celsius",
+                  "type": "boolean"
+                }
+              },
+              "required": [
+                "liquid_name"
+              ],
+              "title": "get_boiling_pointArguments",
+              "type": "object"
+            }
+          }
+        }
+      ]
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-6d7f54b7be48",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": "call_4ldOwO71od1E0lrdgYQCoe2e",
+                    "function": {
+                      "arguments": "",
+                      "name": "get_boiling_point"
+                    },
+                    "type": "function"
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "TdV"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-6d7f54b7be48",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "{\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "L5f"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-6d7f54b7be48",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "li",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "qo3z"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-6d7f54b7be48",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "quid",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "i3"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-6d7f54b7be48",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "_name",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Z"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-6d7f54b7be48",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\":\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Z"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-6d7f54b7be48",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "my",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "QdX5"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-6d7f54b7be48",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "aw",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "sJYi"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-6d7f54b7be48",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "esom",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Yk"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-6d7f54b7be48",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "eli",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "pnS"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-6d7f54b7be48",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "quid",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "y5"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-6d7f54b7be48",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\"}",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Tjs"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-6d7f54b7be48",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": "tool_calls",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Cx0I"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-6d7f54b7be48",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": {
+            "completion_tokens": 22,
+            "prompt_tokens": 156,
+            "total_tokens": 178,
+            "completion_tokens_details": {
+              "accepted_prediction_tokens": 0,
+              "audio_tokens": 0,
+              "reasoning_tokens": 0,
+              "rejected_prediction_tokens": 0
+            },
+            "prompt_tokens_details": {
+              "audio_tokens": 0,
+              "cached_tokens": 0
+            }
+          },
+          "obfuscation": "bmRrd4XLuhmCv"
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/recordings/73c9287059db75cd80dc56cff905fe3ff21e6c39189ab93778335439f288158f.json

@@ -0,0 +1,771 @@
+{
+  "test_id": "tests/integration/responses/test_tool_responses.py::test_response_non_streaming_multi_turn_tool_execution[openai_client-txt=openai/gpt-4o-user_file_access_check]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "I need to check if user 'alice' can access the file 'document.txt'. First, get alice's user ID, then check if that user ID can access the file 'document.txt'. Do this as a series of steps, where each step is a separate message. Return only one tool call per step. Summarize the final result with a single 'yes' or 'no' response."
+        },
+        {
+          "role": "assistant",
+          "content": "",
+          "tool_calls": [
+            {
+              "index": 0,
+              "id": "call_EsVvmBUqtJb42kNkYnK19QkJ",
+              "type": "function",
+              "function": {
+                "name": "get_user_id",
+                "arguments": "{\"username\":\"alice\"}"
+              }
+            }
+          ]
+        },
+        {
+          "role": "tool",
+          "tool_call_id": "call_EsVvmBUqtJb42kNkYnK19QkJ",
+          "content": [
+            {
+              "type": "text",
+              "text": "user_12345"
+            }
+          ]
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      },
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "get_user_id",
+            "description": "\n        Get the user ID for a given username. This ID is needed for other operations.\n\n        :param username: The username to look up\n        :return: The user ID for the username\n        ",
+            "parameters": {
+              "properties": {
+                "username": {
+                  "title": "Username",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "username"
+              ],
+              "title": "get_user_idArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_user_permissions",
+            "description": "\n        Get the permissions for a user ID. Requires a valid user ID from get_user_id.\n\n        :param user_id: The user ID to check permissions for\n        :return: The permissions for the user\n        ",
+            "parameters": {
+              "properties": {
+                "user_id": {
+                  "title": "User Id",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "user_id"
+              ],
+              "title": "get_user_permissionsArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "check_file_access",
+            "description": "\n        Check if a user can access a specific file. Requires a valid user ID.\n\n        :param user_id: The user ID to check access for\n        :param filename: The filename to check access to\n        :return: Whether the user can access the file (yes/no)\n        ",
+            "parameters": {
+              "properties": {
+                "user_id": {
+                  "title": "User Id",
+                  "type": "string"
+                },
+                "filename": {
+                  "title": "Filename",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "user_id",
+                "filename"
+              ],
+              "title": "check_file_accessArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_experiment_id",
+            "description": "\n        Get the experiment ID for a given experiment name. This ID is needed to get results.\n\n        :param experiment_name: The name of the experiment\n        :return: The experiment ID\n        ",
+            "parameters": {
+              "properties": {
+                "experiment_name": {
+                  "title": "Experiment Name",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "experiment_name"
+              ],
+              "title": "get_experiment_idArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_experiment_results",
+            "description": "\n        Get the results for an experiment ID. Requires a valid experiment ID from get_experiment_id.\n\n        :param experiment_id: The experiment ID to get results for\n        :return: The experiment results\n        ",
+            "parameters": {
+              "properties": {
+                "experiment_id": {
+                  "title": "Experiment Id",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "experiment_id"
+              ],
+              "title": "get_experiment_resultsArguments",
+              "type": "object"
+            }
+          }
+        }
+      ]
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-73c9287059db",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": "call_kCmSE8ORKfQoiEsW2UCYr5Sh",
+                    "function": {
+                      "arguments": "",
+                      "name": "check_file_access"
+                    },
+                    "type": "function"
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "sCU"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-73c9287059db",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "{\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "iHp"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-73c9287059db",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "user",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "3b"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-73c9287059db",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "_id",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "4hG"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-73c9287059db",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\":\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "z"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-73c9287059db",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "user",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "zX"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-73c9287059db",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "_",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "WRFf5"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-73c9287059db",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "123",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "PvE"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-73c9287059db",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "45",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "xak8"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-73c9287059db",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\",\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "v"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-73c9287059db",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "filename",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "l7Rfy5le49BJu0"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-73c9287059db",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\":\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "p"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-73c9287059db",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "document",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "EpFPZH128OUIsw"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-73c9287059db",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": ".txt",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Zg"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-73c9287059db",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\"}",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "jH3"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-73c9287059db",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": "tool_calls",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "UubI"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-73c9287059db",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": {
+            "completion_tokens": 24,
+            "prompt_tokens": 482,
+            "total_tokens": 506,
+            "completion_tokens_details": {
+              "accepted_prediction_tokens": 0,
+              "audio_tokens": 0,
+              "reasoning_tokens": 0,
+              "rejected_prediction_tokens": 0
+            },
+            "prompt_tokens_details": {
+              "audio_tokens": 0,
+              "cached_tokens": 0
+            }
+          },
+          "obfuscation": "GITY7sf69sAJd"
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/recordings/775a161a318a252454fd44f9850b37c6ec15eb17dfaa95f015dcc6f65fa10c94.json

@@ -0,0 +1,574 @@
+{
+  "test_id": "tests/integration/responses/test_mcp_authentication.py::test_mcp_authorization_bearer[openai_client-txt=openai/gpt-4o]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "What is the boiling point of myawesomeliquid?"
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      },
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "greet_everyone",
+            "parameters": {
+              "properties": {
+                "url": {
+                  "title": "Url",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "url"
+              ],
+              "title": "greet_everyoneArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_boiling_point",
+            "description": "\n        Returns the boiling point of a liquid in Celsius or Fahrenheit.\n\n        :param liquid_name: The name of the liquid\n        :param celsius: Whether to return the boiling point in Celsius\n        :return: The boiling point of the liquid in Celcius or Fahrenheit\n        ",
+            "parameters": {
+              "properties": {
+                "liquid_name": {
+                  "title": "Liquid Name",
+                  "type": "string"
+                },
+                "celsius": {
+                  "default": true,
+                  "title": "Celsius",
+                  "type": "boolean"
+                }
+              },
+              "required": [
+                "liquid_name"
+              ],
+              "title": "get_boiling_pointArguments",
+              "type": "object"
+            }
+          }
+        }
+      ]
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-775a161a318a",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": "call_2lYntxgdJV66JFvD6OuICQCB",
+                    "function": {
+                      "arguments": "",
+                      "name": "get_boiling_point"
+                    },
+                    "type": "function"
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "UmB"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-775a161a318a",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "{\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "ejb"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-775a161a318a",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "li",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Loxj"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-775a161a318a",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "quid",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "IQ"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-775a161a318a",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "_name",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "8"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-775a161a318a",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\":\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "G"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-775a161a318a",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "my",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "lo9p"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-775a161a318a",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "aw",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "YWPA"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-775a161a318a",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "esom",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "vV"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-775a161a318a",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "eli",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "e0t"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-775a161a318a",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "quid",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "kv"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-775a161a318a",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\"}",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "h2F"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-775a161a318a",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": "tool_calls",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "B9QY"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-775a161a318a",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": {
+            "completion_tokens": 22,
+            "prompt_tokens": 154,
+            "total_tokens": 176,
+            "completion_tokens_details": {
+              "accepted_prediction_tokens": 0,
+              "audio_tokens": 0,
+              "reasoning_tokens": 0,
+              "rejected_prediction_tokens": 0
+            },
+            "prompt_tokens_details": {
+              "audio_tokens": 0,
+              "cached_tokens": 0
+            }
+          },
+          "obfuscation": "MH88zIptmy2Xs"
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/recordings/77ad6e42c34823ac51a784cfe4fa0ee18d09bd413189a7c03b24bf3871e3d8d7.json

@@ -10,7 +10,7 @@
   },
   "response": {
     "body": {
-      "__type__": "llama_stack.apis.tools.tools.ToolInvocationResult",
+      "__type__": "llama_stack_api.tools.ToolInvocationResult",
       "__data__": {
         "content": "{\"query\": \"latest version of Python\", \"top_k\": [{\"url\": \"https://www.liquidweb.com/blog/latest-python-version/\", \"title\": \"The latest Python version: Python 3.14 - Liquid Web\", \"content\": \"The latest major version, Python 3.14 was officially released on October 7, 2025. Let's explore the key features of Python's current version, how to download\", \"score\": 0.890761, \"raw_content\": null}, {\"url\": \"https://docs.python.org/3/whatsnew/3.14.html\", \"title\": \"What's new in Python 3.14 \\u2014 Python 3.14.0 documentation\", \"content\": \"Python 3.14 is the latest stable release of the Python programming language, with a mix of changes to the language, the implementation, and the standard\", \"score\": 0.8124067, \"raw_content\": null}, {\"url\": \"https://devguide.python.org/versions/\", \"title\": \"Status of Python versions - Python Developer's Guide\", \"content\": \"The main branch is currently the future Python 3.15, and is the only branch that accepts new features. The latest release for each Python version can be found\", \"score\": 0.80089486, \"raw_content\": null}, {\"url\": \"https://www.python.org/doc/versions/\", \"title\": \"Python documentation by version\", \"content\": \"Python 3.12.4, documentation released on 6 June 2024. Python 3.12.3, documentation released on 9 April 2024. Python 3.12.2, documentation released on 6 February\", \"score\": 0.74563974, \"raw_content\": null}, {\"url\": \"https://www.python.org/downloads/\", \"title\": \"Download Python | Python.org\", \"content\": \"Active Python Releases \\u00b7 3.15 pre-release 2026-10-07 (planned) 2031-10 PEP 790 \\u00b7 3.14 bugfix 2025-10-07 2030-10 PEP 745 \\u00b7 3.13 bugfix 2024-10-07 2029-10 PEP 719\", \"score\": 0.6551821, \"raw_content\": null}]}",
         "error_message": null,

tests/integration/responses/recordings/9f10c42f1338ae4b535cb877851520db560af78e9bc38159e526b68b8daa168e.json

@@ -0,0 +1,759 @@
+{
+  "test_id": "tests/integration/responses/test_tool_responses.py::test_response_sequential_mcp_tool[openai_client-txt=openai/gpt-4o-boiling_point_tool]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "What is the boiling point of myawesomeliquid in Celsius?"
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      },
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "greet_everyone",
+            "parameters": {
+              "properties": {
+                "url": {
+                  "title": "Url",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "url"
+              ],
+              "title": "greet_everyoneArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_boiling_point",
+            "description": "\n        Returns the boiling point of a liquid in Celsius or Fahrenheit.\n\n        :param liquid_name: The name of the liquid\n        :param celsius: Whether to return the boiling point in Celsius\n        :return: The boiling point of the liquid in Celcius or Fahrenheit\n        ",
+            "parameters": {
+              "properties": {
+                "liquid_name": {
+                  "title": "Liquid Name",
+                  "type": "string"
+                },
+                "celsius": {
+                  "default": true,
+                  "title": "Celsius",
+                  "type": "boolean"
+                }
+              },
+              "required": [
+                "liquid_name"
+              ],
+              "title": "get_boiling_pointArguments",
+              "type": "object"
+            }
+          }
+        }
+      ]
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-9f10c42f1338",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": "call_b5k2yeqIi5ucElnnrVPyYU4x",
+                    "function": {
+                      "arguments": "",
+                      "name": "get_boiling_point"
+                    },
+                    "type": "function"
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "AhH"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-9f10c42f1338",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "{\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "SMa"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-9f10c42f1338",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "li",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "fBD0"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-9f10c42f1338",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "quid",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "LL"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-9f10c42f1338",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "_name",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "h"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-9f10c42f1338",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\":\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "5"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-9f10c42f1338",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "my",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "ySpU"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-9f10c42f1338",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "aw",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "fra1"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-9f10c42f1338",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "esom",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Hb"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-9f10c42f1338",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "eli",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "INi"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-9f10c42f1338",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "quid",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "jF"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-9f10c42f1338",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\",\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "i"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-9f10c42f1338",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "c",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "2dDeK"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-9f10c42f1338",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "elsius",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": ""
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-9f10c42f1338",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\":",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "DSb"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-9f10c42f1338",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "true",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "vP"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-9f10c42f1338",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "}",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "9boiy"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-9f10c42f1338",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": "tool_calls",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "ZZRa"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-9f10c42f1338",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": {
+            "completion_tokens": 27,
+            "prompt_tokens": 156,
+            "total_tokens": 183,
+            "completion_tokens_details": {
+              "accepted_prediction_tokens": 0,
+              "audio_tokens": 0,
+              "reasoning_tokens": 0,
+              "rejected_prediction_tokens": 0
+            },
+            "prompt_tokens_details": {
+              "audio_tokens": 0,
+              "cached_tokens": 0
+            }
+          },
+          "obfuscation": "HoutUcx6gZI1g"
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/recordings/a97d8a2f2fd75b4a5ca732e632b981ca011dd1b6c29df530d12726b1cf7989e5.json

@@ -0,0 +1,833 @@
+{
+  "test_id": "tests/integration/responses/test_tool_responses.py::test_response_streaming_multi_turn_tool_execution[openai_client-txt=openai/gpt-4o-user_permissions_workflow]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "Help me with this security check: First, get the user ID for 'charlie', then get the permissions for that user ID, and finally check if that user can access 'secret_file.txt'. Stream your progress as you work through each step. Return only one tool call per step. Summarize the final result with a single 'yes' or 'no' response."
+        },
+        {
+          "role": "assistant",
+          "content": "",
+          "tool_calls": [
+            {
+              "index": 0,
+              "id": "call_fsxGbKmceUbLSXCe4sx9WLXO",
+              "type": "function",
+              "function": {
+                "name": "get_user_id",
+                "arguments": "{\"username\":\"charlie\"}"
+              }
+            }
+          ]
+        },
+        {
+          "role": "tool",
+          "tool_call_id": "call_fsxGbKmceUbLSXCe4sx9WLXO",
+          "content": [
+            {
+              "type": "text",
+              "text": "user_11111"
+            }
+          ]
+        },
+        {
+          "role": "assistant",
+          "content": "",
+          "tool_calls": [
+            {
+              "index": 0,
+              "id": "call_moRBxqnBJ48EWTSEoQ1llgib",
+              "type": "function",
+              "function": {
+                "name": "get_user_permissions",
+                "arguments": "{\"user_id\":\"user_11111\"}"
+              }
+            }
+          ]
+        },
+        {
+          "role": "tool",
+          "tool_call_id": "call_moRBxqnBJ48EWTSEoQ1llgib",
+          "content": [
+            {
+              "type": "text",
+              "text": "admin"
+            }
+          ]
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      },
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "get_user_id",
+            "description": "\n        Get the user ID for a given username. This ID is needed for other operations.\n\n        :param username: The username to look up\n        :return: The user ID for the username\n        ",
+            "parameters": {
+              "properties": {
+                "username": {
+                  "title": "Username",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "username"
+              ],
+              "title": "get_user_idArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_user_permissions",
+            "description": "\n        Get the permissions for a user ID. Requires a valid user ID from get_user_id.\n\n        :param user_id: The user ID to check permissions for\n        :return: The permissions for the user\n        ",
+            "parameters": {
+              "properties": {
+                "user_id": {
+                  "title": "User Id",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "user_id"
+              ],
+              "title": "get_user_permissionsArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "check_file_access",
+            "description": "\n        Check if a user can access a specific file. Requires a valid user ID.\n\n        :param user_id: The user ID to check access for\n        :param filename: The filename to check access to\n        :return: Whether the user can access the file (yes/no)\n        ",
+            "parameters": {
+              "properties": {
+                "user_id": {
+                  "title": "User Id",
+                  "type": "string"
+                },
+                "filename": {
+                  "title": "Filename",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "user_id",
+                "filename"
+              ],
+              "title": "check_file_accessArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_experiment_id",
+            "description": "\n        Get the experiment ID for a given experiment name. This ID is needed to get results.\n\n        :param experiment_name: The name of the experiment\n        :return: The experiment ID\n        ",
+            "parameters": {
+              "properties": {
+                "experiment_name": {
+                  "title": "Experiment Name",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "experiment_name"
+              ],
+              "title": "get_experiment_idArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_experiment_results",
+            "description": "\n        Get the results for an experiment ID. Requires a valid experiment ID from get_experiment_id.\n\n        :param experiment_id: The experiment ID to get results for\n        :return: The experiment results\n        ",
+            "parameters": {
+              "properties": {
+                "experiment_id": {
+                  "title": "Experiment Id",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "experiment_id"
+              ],
+              "title": "get_experiment_resultsArguments",
+              "type": "object"
+            }
+          }
+        }
+      ]
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-a97d8a2f2fd7",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": "call_ybUqAP9oQn3rwQqVdOLs5Wb4",
+                    "function": {
+                      "arguments": "",
+                      "name": "check_file_access"
+                    },
+                    "type": "function"
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "xpc"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-a97d8a2f2fd7",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "{\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "xXs"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-a97d8a2f2fd7",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "user",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "XY"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-a97d8a2f2fd7",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "_id",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "HbC"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-a97d8a2f2fd7",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\":\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "f"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-a97d8a2f2fd7",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "user",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Ds"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-a97d8a2f2fd7",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "_",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Osfy3"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-a97d8a2f2fd7",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "111",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "ioI"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-a97d8a2f2fd7",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "11",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "GQg6"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-a97d8a2f2fd7",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\",\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "2"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-a97d8a2f2fd7",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "filename",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "b2qqKbGC68nHMB"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-a97d8a2f2fd7",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\":\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "H"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-a97d8a2f2fd7",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "secret",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": ""
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-a97d8a2f2fd7",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "_file",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "6"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-a97d8a2f2fd7",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": ".txt",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Wz"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-a97d8a2f2fd7",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\"}",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "ImW"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-a97d8a2f2fd7",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": "tool_calls",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "nRAE"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-a97d8a2f2fd7",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": {
+            "completion_tokens": 25,
+            "prompt_tokens": 507,
+            "total_tokens": 532,
+            "completion_tokens_details": {
+              "accepted_prediction_tokens": 0,
+              "audio_tokens": 0,
+              "reasoning_tokens": 0,
+              "rejected_prediction_tokens": 0
+            },
+            "prompt_tokens_details": {
+              "audio_tokens": 0,
+              "cached_tokens": 0
+            }
+          },
+          "obfuscation": "rgbYyZ54cN8La"
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/recordings/b30da63114770b8c975bf66e24aee40546a0658db3df58b9b4d948e4e95b0961.json

@@ -0,0 +1,524 @@
+{
+  "test_id": "tests/integration/responses/test_tool_responses.py::test_response_streaming_multi_turn_tool_execution[openai_client-txt=openai/gpt-4o-experiment_analysis_streaming]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "I need a complete analysis: First, get the experiment ID for 'chemical_reaction', then get the results for that experiment, and tell me if the yield was above 80%. Return only one tool call per step.  Please stream your analysis process."
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      },
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "get_user_id",
+            "description": "\n        Get the user ID for a given username. This ID is needed for other operations.\n\n        :param username: The username to look up\n        :return: The user ID for the username\n        ",
+            "parameters": {
+              "properties": {
+                "username": {
+                  "title": "Username",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "username"
+              ],
+              "title": "get_user_idArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_user_permissions",
+            "description": "\n        Get the permissions for a user ID. Requires a valid user ID from get_user_id.\n\n        :param user_id: The user ID to check permissions for\n        :return: The permissions for the user\n        ",
+            "parameters": {
+              "properties": {
+                "user_id": {
+                  "title": "User Id",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "user_id"
+              ],
+              "title": "get_user_permissionsArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "check_file_access",
+            "description": "\n        Check if a user can access a specific file. Requires a valid user ID.\n\n        :param user_id: The user ID to check access for\n        :param filename: The filename to check access to\n        :return: Whether the user can access the file (yes/no)\n        ",
+            "parameters": {
+              "properties": {
+                "user_id": {
+                  "title": "User Id",
+                  "type": "string"
+                },
+                "filename": {
+                  "title": "Filename",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "user_id",
+                "filename"
+              ],
+              "title": "check_file_accessArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_experiment_id",
+            "description": "\n        Get the experiment ID for a given experiment name. This ID is needed to get results.\n\n        :param experiment_name: The name of the experiment\n        :return: The experiment ID\n        ",
+            "parameters": {
+              "properties": {
+                "experiment_name": {
+                  "title": "Experiment Name",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "experiment_name"
+              ],
+              "title": "get_experiment_idArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_experiment_results",
+            "description": "\n        Get the results for an experiment ID. Requires a valid experiment ID from get_experiment_id.\n\n        :param experiment_id: The experiment ID to get results for\n        :return: The experiment results\n        ",
+            "parameters": {
+              "properties": {
+                "experiment_id": {
+                  "title": "Experiment Id",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "experiment_id"
+              ],
+              "title": "get_experiment_resultsArguments",
+              "type": "object"
+            }
+          }
+        }
+      ]
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-b30da6311477",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": "call_Q9Gcxub7UbQsxJWVkiy4FETr",
+                    "function": {
+                      "arguments": "",
+                      "name": "get_experiment_id"
+                    },
+                    "type": "function"
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "c8d"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-b30da6311477",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "{\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "QoE"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-b30da6311477",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "experiment",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "1krtmewG8p36"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-b30da6311477",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "_name",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "P"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-b30da6311477",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\":\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "D"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-b30da6311477",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "chemical",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "FoS4ov7pi99K5h"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-b30da6311477",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "_re",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "BhD"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-b30da6311477",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "action",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": ""
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-b30da6311477",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\"}",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "KWC"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-b30da6311477",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": "tool_calls",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "PFmv"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-b30da6311477",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": {
+            "completion_tokens": 19,
+            "prompt_tokens": 425,
+            "total_tokens": 444,
+            "completion_tokens_details": {
+              "accepted_prediction_tokens": 0,
+              "audio_tokens": 0,
+              "reasoning_tokens": 0,
+              "rejected_prediction_tokens": 0
+            },
+            "prompt_tokens_details": {
+              "audio_tokens": 0,
+              "cached_tokens": 0
+            }
+          },
+          "obfuscation": "NYdC3zepOXLsO"
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/recordings/b6b7282ca0ad5a3c59321d2b045a91ebca1cbaeb4f7aab22c5b9e246b476272f.json

@@ -0,0 +1,649 @@
+{
+  "test_id": "tests/integration/responses/test_tool_responses.py::test_response_sequential_mcp_tool[openai_client-txt=openai/gpt-4o-boiling_point_tool]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "What is the boiling point of myawesomeliquid in Celsius?"
+        },
+        {
+          "role": "assistant",
+          "content": "",
+          "tool_calls": [
+            {
+              "index": 0,
+              "id": "call_b5k2yeqIi5ucElnnrVPyYU4x",
+              "type": "function",
+              "function": {
+                "name": "get_boiling_point",
+                "arguments": "{\"liquid_name\":\"myawesomeliquid\",\"celsius\":true}"
+              }
+            }
+          ]
+        },
+        {
+          "role": "tool",
+          "tool_call_id": "call_b5k2yeqIi5ucElnnrVPyYU4x",
+          "content": [
+            {
+              "type": "text",
+              "text": "-100"
+            }
+          ]
+        },
+        {
+          "role": "assistant",
+          "content": "The boiling point of \"myawesomeliquid\" is -100 degrees Celsius."
+        },
+        {
+          "role": "user",
+          "content": "What is the boiling point of myawesomeliquid in Celsius?"
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      },
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "greet_everyone",
+            "parameters": {
+              "properties": {
+                "url": {
+                  "title": "Url",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "url"
+              ],
+              "title": "greet_everyoneArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_boiling_point",
+            "description": "\n        Returns the boiling point of a liquid in Celsius or Fahrenheit.\n\n        :param liquid_name: The name of the liquid\n        :param celsius: Whether to return the boiling point in Celsius\n        :return: The boiling point of the liquid in Celcius or Fahrenheit\n        ",
+            "parameters": {
+              "properties": {
+                "liquid_name": {
+                  "title": "Liquid Name",
+                  "type": "string"
+                },
+                "celsius": {
+                  "default": true,
+                  "title": "Celsius",
+                  "type": "boolean"
+                }
+              },
+              "required": [
+                "liquid_name"
+              ],
+              "title": "get_boiling_pointArguments",
+              "type": "object"
+            }
+          }
+        }
+      ]
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-b6b7282ca0ad",
+          "choices": [
+            {
+              "delta": {
+                "content": "",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "7S5XpbMeFTTZba"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-b6b7282ca0ad",
+          "choices": [
+            {
+              "delta": {
+                "content": "The",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "G4KYajpQCgm5p"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-b6b7282ca0ad",
+          "choices": [
+            {
+              "delta": {
+                "content": " boiling",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "krw8d3Np"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-b6b7282ca0ad",
+          "choices": [
+            {
+              "delta": {
+                "content": " point",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "sOEsvVtCEV"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-b6b7282ca0ad",
+          "choices": [
+            {
+              "delta": {
+                "content": " of",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "5eAw89OUrx7VT"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-b6b7282ca0ad",
+          "choices": [
+            {
+              "delta": {
+                "content": " \"",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "PFghmTocqCYea"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-b6b7282ca0ad",
+          "choices": [
+            {
+              "delta": {
+                "content": "my",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "IRJRbKIoXwNh0e"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-b6b7282ca0ad",
+          "choices": [
+            {
+              "delta": {
+                "content": "aw",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "wuoL6MoA21KfMP"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-b6b7282ca0ad",
+          "choices": [
+            {
+              "delta": {
+                "content": "esom",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "DLRS3D5YVekk"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-b6b7282ca0ad",
+          "choices": [
+            {
+              "delta": {
+                "content": "eli",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "PQZQlOncwl01F"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-b6b7282ca0ad",
+          "choices": [
+            {
+              "delta": {
+                "content": "quid",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "TVfNNxYtZgXQ"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-b6b7282ca0ad",
+          "choices": [
+            {
+              "delta": {
+                "content": "\"",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "LscPqJGnbMf6Qw"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-b6b7282ca0ad",
+          "choices": [
+            {
+              "delta": {
+                "content": " is",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "X8NSrxHcpYYXL"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-b6b7282ca0ad",
+          "choices": [
+            {
+              "delta": {
+                "content": " -",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "5nfdb4DuFapoeT"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-b6b7282ca0ad",
+          "choices": [
+            {
+              "delta": {
+                "content": "100",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "K2qXQYFAd591w"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-b6b7282ca0ad",
+          "choices": [
+            {
+              "delta": {
+                "content": " degrees",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "b0rvHdF1"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-b6b7282ca0ad",
+          "choices": [
+            {
+              "delta": {
+                "content": " Celsius",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "kFoGt52c"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-b6b7282ca0ad",
+          "choices": [
+            {
+              "delta": {
+                "content": ".",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "SJjhJwz2zgz693C"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-b6b7282ca0ad",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": "stop",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "MityMxFgBz"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-b6b7282ca0ad",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": {
+            "completion_tokens": 18,
+            "prompt_tokens": 234,
+            "total_tokens": 252,
+            "completion_tokens_details": {
+              "accepted_prediction_tokens": 0,
+              "audio_tokens": 0,
+              "reasoning_tokens": 0,
+              "rejected_prediction_tokens": 0
+            },
+            "prompt_tokens_details": {
+              "audio_tokens": 0,
+              "cached_tokens": 0
+            }
+          },
+          "obfuscation": "qf0j6dzuNPifV"
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/recordings/c27df465b2996c4d7c909e9ccfac53deb1ac47d064a1b5c70a78b7436438818f.json

@@ -0,0 +1,450 @@
+{
+  "test_id": "tests/integration/responses/test_tool_responses.py::test_response_streaming_multi_turn_tool_execution[openai_client-txt=openai/gpt-4o-user_permissions_workflow]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "Help me with this security check: First, get the user ID for 'charlie', then get the permissions for that user ID, and finally check if that user can access 'secret_file.txt'. Stream your progress as you work through each step. Return only one tool call per step. Summarize the final result with a single 'yes' or 'no' response."
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      },
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "get_user_id",
+            "description": "\n        Get the user ID for a given username. This ID is needed for other operations.\n\n        :param username: The username to look up\n        :return: The user ID for the username\n        ",
+            "parameters": {
+              "properties": {
+                "username": {
+                  "title": "Username",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "username"
+              ],
+              "title": "get_user_idArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_user_permissions",
+            "description": "\n        Get the permissions for a user ID. Requires a valid user ID from get_user_id.\n\n        :param user_id: The user ID to check permissions for\n        :return: The permissions for the user\n        ",
+            "parameters": {
+              "properties": {
+                "user_id": {
+                  "title": "User Id",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "user_id"
+              ],
+              "title": "get_user_permissionsArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "check_file_access",
+            "description": "\n        Check if a user can access a specific file. Requires a valid user ID.\n\n        :param user_id: The user ID to check access for\n        :param filename: The filename to check access to\n        :return: Whether the user can access the file (yes/no)\n        ",
+            "parameters": {
+              "properties": {
+                "user_id": {
+                  "title": "User Id",
+                  "type": "string"
+                },
+                "filename": {
+                  "title": "Filename",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "user_id",
+                "filename"
+              ],
+              "title": "check_file_accessArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_experiment_id",
+            "description": "\n        Get the experiment ID for a given experiment name. This ID is needed to get results.\n\n        :param experiment_name: The name of the experiment\n        :return: The experiment ID\n        ",
+            "parameters": {
+              "properties": {
+                "experiment_name": {
+                  "title": "Experiment Name",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "experiment_name"
+              ],
+              "title": "get_experiment_idArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_experiment_results",
+            "description": "\n        Get the results for an experiment ID. Requires a valid experiment ID from get_experiment_id.\n\n        :param experiment_id: The experiment ID to get results for\n        :return: The experiment results\n        ",
+            "parameters": {
+              "properties": {
+                "experiment_id": {
+                  "title": "Experiment Id",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "experiment_id"
+              ],
+              "title": "get_experiment_resultsArguments",
+              "type": "object"
+            }
+          }
+        }
+      ]
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c27df465b299",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": "call_fsxGbKmceUbLSXCe4sx9WLXO",
+                    "function": {
+                      "arguments": "",
+                      "name": "get_user_id"
+                    },
+                    "type": "function"
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "sOa6fZEKZ"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c27df465b299",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "{\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "HBO"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c27df465b299",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "username",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "7kcXlaglccmA8a"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c27df465b299",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\":\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "a"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c27df465b299",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "char",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "bS"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c27df465b299",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "lie",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "d2e"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c27df465b299",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\"}",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "fhE"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c27df465b299",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": "tool_calls",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "SlsZ"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c27df465b299",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": {
+            "completion_tokens": 16,
+            "prompt_tokens": 449,
+            "total_tokens": 465,
+            "completion_tokens_details": {
+              "accepted_prediction_tokens": 0,
+              "audio_tokens": 0,
+              "reasoning_tokens": 0,
+              "rejected_prediction_tokens": 0
+            },
+            "prompt_tokens_details": {
+              "audio_tokens": 0,
+              "cached_tokens": 0
+            }
+          },
+          "obfuscation": "fjMWRTbF1Ni06"
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/recordings/c84e894f47a6d7f4d4556829d24ea14cd2869c77972c33e66d9b42438e2165cd.json

@@ -0,0 +1,574 @@
+{
+  "test_id": "tests/integration/responses/test_mcp_authentication.py::test_mcp_authorization_bearer[client_with_models-txt=openai/gpt-4o]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "What is the boiling point of myawesomeliquid?"
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      },
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "greet_everyone",
+            "parameters": {
+              "properties": {
+                "url": {
+                  "title": "Url",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "url"
+              ],
+              "title": "greet_everyoneArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_boiling_point",
+            "description": "\n        Returns the boiling point of a liquid in Celsius or Fahrenheit.\n\n        :param liquid_name: The name of the liquid\n        :param celsius: Whether to return the boiling point in Celsius\n        :return: The boiling point of the liquid in Celcius or Fahrenheit\n        ",
+            "parameters": {
+              "properties": {
+                "liquid_name": {
+                  "title": "Liquid Name",
+                  "type": "string"
+                },
+                "celsius": {
+                  "default": true,
+                  "title": "Celsius",
+                  "type": "boolean"
+                }
+              },
+              "required": [
+                "liquid_name"
+              ],
+              "title": "get_boiling_pointArguments",
+              "type": "object"
+            }
+          }
+        }
+      ]
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c84e894f47a6",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": "call_mitVYvmPaFfoSmKjzKo5xmZp",
+                    "function": {
+                      "arguments": "",
+                      "name": "get_boiling_point"
+                    },
+                    "type": "function"
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "5Y1"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c84e894f47a6",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "{\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "QzQ"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c84e894f47a6",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "li",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "4NPm"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c84e894f47a6",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "quid",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Lh"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c84e894f47a6",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "_name",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "r"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c84e894f47a6",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\":\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "w"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c84e894f47a6",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "my",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "GSVa"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c84e894f47a6",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "aw",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "AWZm"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c84e894f47a6",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "esom",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "DG"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c84e894f47a6",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "eli",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "1Bw"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c84e894f47a6",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "quid",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Oq"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c84e894f47a6",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\"}",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "cI8"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c84e894f47a6",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": "tool_calls",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "kKqh"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c84e894f47a6",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": {
+            "completion_tokens": 22,
+            "prompt_tokens": 154,
+            "total_tokens": 176,
+            "completion_tokens_details": {
+              "accepted_prediction_tokens": 0,
+              "audio_tokens": 0,
+              "reasoning_tokens": 0,
+              "rejected_prediction_tokens": 0
+            },
+            "prompt_tokens_details": {
+              "audio_tokens": 0,
+              "cached_tokens": 0
+            }
+          },
+          "obfuscation": "etTUytEvlkJ99"
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/recordings/c9c723cd01233311d9033f55d6db610b38555bb86f93c507ede8752af47cda6a.json

@@ -0,0 +1,574 @@
+{
+  "test_id": "tests/integration/responses/test_mcp_authentication.py::test_mcp_authorization_backward_compatibility[client_with_models-txt=openai/gpt-4o]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "What is the boiling point of myawesomeliquid?"
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      },
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "greet_everyone",
+            "parameters": {
+              "properties": {
+                "url": {
+                  "title": "Url",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "url"
+              ],
+              "title": "greet_everyoneArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_boiling_point",
+            "description": "\n        Returns the boiling point of a liquid in Celsius or Fahrenheit.\n\n        :param liquid_name: The name of the liquid\n        :param celsius: Whether to return the boiling point in Celsius\n        :return: The boiling point of the liquid in Celcius or Fahrenheit\n        ",
+            "parameters": {
+              "properties": {
+                "liquid_name": {
+                  "title": "Liquid Name",
+                  "type": "string"
+                },
+                "celsius": {
+                  "default": true,
+                  "title": "Celsius",
+                  "type": "boolean"
+                }
+              },
+              "required": [
+                "liquid_name"
+              ],
+              "title": "get_boiling_pointArguments",
+              "type": "object"
+            }
+          }
+        }
+      ]
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c9c723cd0123",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": "call_wnbihJuwYAfnI8uxy84Yl48j",
+                    "function": {
+                      "arguments": "",
+                      "name": "get_boiling_point"
+                    },
+                    "type": "function"
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "TC0"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c9c723cd0123",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "{\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "hDL"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c9c723cd0123",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "li",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "4G8Z"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c9c723cd0123",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "quid",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "ow"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c9c723cd0123",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "_name",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "P"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c9c723cd0123",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\":\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "M"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c9c723cd0123",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "my",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "yhAk"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c9c723cd0123",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "aw",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "SdIN"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c9c723cd0123",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "esom",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "2z"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c9c723cd0123",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "eli",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "nEC"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c9c723cd0123",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "quid",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "2B"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c9c723cd0123",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\"}",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "DoL"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c9c723cd0123",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": "tool_calls",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "cSRf"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-c9c723cd0123",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": {
+            "completion_tokens": 22,
+            "prompt_tokens": 154,
+            "total_tokens": 176,
+            "completion_tokens_details": {
+              "accepted_prediction_tokens": 0,
+              "audio_tokens": 0,
+              "reasoning_tokens": 0,
+              "rejected_prediction_tokens": 0
+            },
+            "prompt_tokens_details": {
+              "audio_tokens": 0,
+              "cached_tokens": 0
+            }
+          },
+          "obfuscation": "ejlSF0NzXFFso"
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/recordings/d35c1244fbbe9898da3958113c1d054d5f5dd6bdd3c4333db6cef7361fb32feb.json

@@ -0,0 +1,759 @@
+{
+  "test_id": "tests/integration/responses/test_tool_responses.py::test_response_non_streaming_mcp_tool[openai_client-txt=openai/gpt-4o-boiling_point_tool]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "What is the boiling point of myawesomeliquid in Celsius?"
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      },
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "greet_everyone",
+            "parameters": {
+              "properties": {
+                "url": {
+                  "title": "Url",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "url"
+              ],
+              "title": "greet_everyoneArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_boiling_point",
+            "description": "\n        Returns the boiling point of a liquid in Celsius or Fahrenheit.\n\n        :param liquid_name: The name of the liquid\n        :param celsius: Whether to return the boiling point in Celsius\n        :return: The boiling point of the liquid in Celcius or Fahrenheit\n        ",
+            "parameters": {
+              "properties": {
+                "liquid_name": {
+                  "title": "Liquid Name",
+                  "type": "string"
+                },
+                "celsius": {
+                  "default": true,
+                  "title": "Celsius",
+                  "type": "boolean"
+                }
+              },
+              "required": [
+                "liquid_name"
+              ],
+              "title": "get_boiling_pointArguments",
+              "type": "object"
+            }
+          }
+        }
+      ]
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d35c1244fbbe",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": "call_8kf8fNIDcWOelbCmUEcretON",
+                    "function": {
+                      "arguments": "",
+                      "name": "get_boiling_point"
+                    },
+                    "type": "function"
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "1xG"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d35c1244fbbe",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "{\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "RQj"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d35c1244fbbe",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "li",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "XncI"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d35c1244fbbe",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "quid",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "86"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d35c1244fbbe",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "_name",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "L"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d35c1244fbbe",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\":\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "8"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d35c1244fbbe",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "my",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "lnSu"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d35c1244fbbe",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "aw",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "ksr1"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d35c1244fbbe",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "esom",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "CU"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d35c1244fbbe",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "eli",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "hrv"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d35c1244fbbe",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "quid",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "K9"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d35c1244fbbe",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\",\"",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "a"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d35c1244fbbe",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "c",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "LKw52"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d35c1244fbbe",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "elsius",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": ""
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d35c1244fbbe",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "\":",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "yGY"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d35c1244fbbe",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "true",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "wC"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d35c1244fbbe",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": null,
+                    "function": {
+                      "arguments": "}",
+                      "name": null
+                    },
+                    "type": null
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "8fF8B"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d35c1244fbbe",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": "tool_calls",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "bbwp"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d35c1244fbbe",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": {
+            "completion_tokens": 27,
+            "prompt_tokens": 156,
+            "total_tokens": 183,
+            "completion_tokens_details": {
+              "accepted_prediction_tokens": 0,
+              "audio_tokens": 0,
+              "reasoning_tokens": 0,
+              "rejected_prediction_tokens": 0
+            },
+            "prompt_tokens_details": {
+              "audio_tokens": 0,
+              "cached_tokens": 0
+            }
+          },
+          "obfuscation": "k0bo4JwUfLNKW"
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/recordings/d42e1020edee86d9f6da7df909c2a453cb8f2e11e80beb8e5506439345c428eb.json

@@ -0,0 +1,808 @@
+{
+  "test_id": "tests/integration/responses/test_tool_responses.py::test_response_streaming_multi_turn_tool_execution[openai_client-txt=openai/gpt-4o-experiment_analysis_streaming]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "I need a complete analysis: First, get the experiment ID for 'chemical_reaction', then get the results for that experiment, and tell me if the yield was above 80%. Return only one tool call per step.  Please stream your analysis process."
+        },
+        {
+          "role": "assistant",
+          "content": "",
+          "tool_calls": [
+            {
+              "index": 0,
+              "id": "call_Q9Gcxub7UbQsxJWVkiy4FETr",
+              "type": "function",
+              "function": {
+                "name": "get_experiment_id",
+                "arguments": "{\"experiment_name\":\"chemical_reaction\"}"
+              }
+            }
+          ]
+        },
+        {
+          "role": "tool",
+          "tool_call_id": "call_Q9Gcxub7UbQsxJWVkiy4FETr",
+          "content": [
+            {
+              "type": "text",
+              "text": "exp_003"
+            }
+          ]
+        },
+        {
+          "role": "assistant",
+          "content": "",
+          "tool_calls": [
+            {
+              "index": 0,
+              "id": "call_yTMuQEKu7x115q8XvhqelRub",
+              "type": "function",
+              "function": {
+                "name": "get_experiment_results",
+                "arguments": "{\"experiment_id\":\"exp_003\"}"
+              }
+            }
+          ]
+        },
+        {
+          "role": "tool",
+          "tool_call_id": "call_yTMuQEKu7x115q8XvhqelRub",
+          "content": [
+            {
+              "type": "text",
+              "text": "Yield: 85%, Status: Complete"
+            }
+          ]
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      },
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "get_user_id",
+            "description": "\n        Get the user ID for a given username. This ID is needed for other operations.\n\n        :param username: The username to look up\n        :return: The user ID for the username\n        ",
+            "parameters": {
+              "properties": {
+                "username": {
+                  "title": "Username",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "username"
+              ],
+              "title": "get_user_idArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_user_permissions",
+            "description": "\n        Get the permissions for a user ID. Requires a valid user ID from get_user_id.\n\n        :param user_id: The user ID to check permissions for\n        :return: The permissions for the user\n        ",
+            "parameters": {
+              "properties": {
+                "user_id": {
+                  "title": "User Id",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "user_id"
+              ],
+              "title": "get_user_permissionsArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "check_file_access",
+            "description": "\n        Check if a user can access a specific file. Requires a valid user ID.\n\n        :param user_id: The user ID to check access for\n        :param filename: The filename to check access to\n        :return: Whether the user can access the file (yes/no)\n        ",
+            "parameters": {
+              "properties": {
+                "user_id": {
+                  "title": "User Id",
+                  "type": "string"
+                },
+                "filename": {
+                  "title": "Filename",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "user_id",
+                "filename"
+              ],
+              "title": "check_file_accessArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_experiment_id",
+            "description": "\n        Get the experiment ID for a given experiment name. This ID is needed to get results.\n\n        :param experiment_name: The name of the experiment\n        :return: The experiment ID\n        ",
+            "parameters": {
+              "properties": {
+                "experiment_name": {
+                  "title": "Experiment Name",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "experiment_name"
+              ],
+              "title": "get_experiment_idArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_experiment_results",
+            "description": "\n        Get the results for an experiment ID. Requires a valid experiment ID from get_experiment_id.\n\n        :param experiment_id: The experiment ID to get results for\n        :return: The experiment results\n        ",
+            "parameters": {
+              "properties": {
+                "experiment_id": {
+                  "title": "Experiment Id",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "experiment_id"
+              ],
+              "title": "get_experiment_resultsArguments",
+              "type": "object"
+            }
+          }
+        }
+      ]
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d42e1020edee",
+          "choices": [
+            {
+              "delta": {
+                "content": "",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "7yA3503fehs27D"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d42e1020edee",
+          "choices": [
+            {
+              "delta": {
+                "content": "The",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "T95BeWrgJQMHt"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d42e1020edee",
+          "choices": [
+            {
+              "delta": {
+                "content": " yield",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "VveNEnHuMQ"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d42e1020edee",
+          "choices": [
+            {
+              "delta": {
+                "content": " for",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "KupSssWahehO"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d42e1020edee",
+          "choices": [
+            {
+              "delta": {
+                "content": " the",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Ogot8KLW0IXw"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d42e1020edee",
+          "choices": [
+            {
+              "delta": {
+                "content": " '",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "dYKJ6jPstuAso4"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d42e1020edee",
+          "choices": [
+            {
+              "delta": {
+                "content": "chemical",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "wcSKhZVd"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d42e1020edee",
+          "choices": [
+            {
+              "delta": {
+                "content": "_re",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "6ZlTlRGLyclHo"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d42e1020edee",
+          "choices": [
+            {
+              "delta": {
+                "content": "action",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "WpYqOmrhXr"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d42e1020edee",
+          "choices": [
+            {
+              "delta": {
+                "content": "'",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "qUhq7HrrwdFEyuY"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d42e1020edee",
+          "choices": [
+            {
+              "delta": {
+                "content": " experiment",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "WWO2y"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d42e1020edee",
+          "choices": [
+            {
+              "delta": {
+                "content": " is",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "pFVMO1BRN37n4"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d42e1020edee",
+          "choices": [
+            {
+              "delta": {
+                "content": " ",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "TtQlcHeU2mPl830"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d42e1020edee",
+          "choices": [
+            {
+              "delta": {
+                "content": "85",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "zyw8OdA0pXZCp5"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d42e1020edee",
+          "choices": [
+            {
+              "delta": {
+                "content": "%,",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "VcHVTGGXrqvev1"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d42e1020edee",
+          "choices": [
+            {
+              "delta": {
+                "content": " which",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "FI9FAA2rX6"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d42e1020edee",
+          "choices": [
+            {
+              "delta": {
+                "content": " is",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Cc65gPYGA6Xfd"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d42e1020edee",
+          "choices": [
+            {
+              "delta": {
+                "content": " above",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "T7BlLMIQGs"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d42e1020edee",
+          "choices": [
+            {
+              "delta": {
+                "content": " ",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "2oKThCybRdG8MzZ"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d42e1020edee",
+          "choices": [
+            {
+              "delta": {
+                "content": "80",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "QHWdJWXK6hzQVS"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d42e1020edee",
+          "choices": [
+            {
+              "delta": {
+                "content": "%.",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "lJnplmQYyl0SL3"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d42e1020edee",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": "stop",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "NPaAVrOB4J"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d42e1020edee",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": {
+            "completion_tokens": 21,
+            "prompt_tokens": 494,
+            "total_tokens": 515,
+            "completion_tokens_details": {
+              "accepted_prediction_tokens": 0,
+              "audio_tokens": 0,
+              "reasoning_tokens": 0,
+              "rejected_prediction_tokens": 0
+            },
+            "prompt_tokens_details": {
+              "audio_tokens": 0,
+              "cached_tokens": 0
+            }
+          },
+          "obfuscation": "ngidabPDDHECm"
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/recordings/db81127157a8364ce8f7a81e10d9b84bf814950e3c8f11eed7ed9f11d4462237.json

@@ -0,0 +1,614 @@
+{
+  "test_id": "tests/integration/responses/test_mcp_authentication.py::test_mcp_authorization_backward_compatibility[client_with_models-txt=openai/gpt-4o]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "What is the boiling point of myawesomeliquid?"
+        },
+        {
+          "role": "assistant",
+          "content": "",
+          "tool_calls": [
+            {
+              "index": 0,
+              "id": "call_wnbihJuwYAfnI8uxy84Yl48j",
+              "type": "function",
+              "function": {
+                "name": "get_boiling_point",
+                "arguments": "{\"liquid_name\":\"myawesomeliquid\"}"
+              }
+            }
+          ]
+        },
+        {
+          "role": "tool",
+          "tool_call_id": "call_wnbihJuwYAfnI8uxy84Yl48j",
+          "content": [
+            {
+              "type": "text",
+              "text": "-100"
+            }
+          ]
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      },
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "greet_everyone",
+            "parameters": {
+              "properties": {
+                "url": {
+                  "title": "Url",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "url"
+              ],
+              "title": "greet_everyoneArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_boiling_point",
+            "description": "\n        Returns the boiling point of a liquid in Celsius or Fahrenheit.\n\n        :param liquid_name: The name of the liquid\n        :param celsius: Whether to return the boiling point in Celsius\n        :return: The boiling point of the liquid in Celcius or Fahrenheit\n        ",
+            "parameters": {
+              "properties": {
+                "liquid_name": {
+                  "title": "Liquid Name",
+                  "type": "string"
+                },
+                "celsius": {
+                  "default": true,
+                  "title": "Celsius",
+                  "type": "boolean"
+                }
+              },
+              "required": [
+                "liquid_name"
+              ],
+              "title": "get_boiling_pointArguments",
+              "type": "object"
+            }
+          }
+        }
+      ]
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-db81127157a8",
+          "choices": [
+            {
+              "delta": {
+                "content": "",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Usdowqbd6beiYB"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-db81127157a8",
+          "choices": [
+            {
+              "delta": {
+                "content": "The",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "nVevItSH27TBR"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-db81127157a8",
+          "choices": [
+            {
+              "delta": {
+                "content": " boiling",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "HWyYtVAl"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-db81127157a8",
+          "choices": [
+            {
+              "delta": {
+                "content": " point",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "kvvcut6Eib"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-db81127157a8",
+          "choices": [
+            {
+              "delta": {
+                "content": " of",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "E0osAbGBpCPvy"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-db81127157a8",
+          "choices": [
+            {
+              "delta": {
+                "content": " \"",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "GmH7m44fmv0Mk"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-db81127157a8",
+          "choices": [
+            {
+              "delta": {
+                "content": "my",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "oJ4DV7z5GiqJqX"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-db81127157a8",
+          "choices": [
+            {
+              "delta": {
+                "content": "aw",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "8AmNNAYPXMNrEr"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-db81127157a8",
+          "choices": [
+            {
+              "delta": {
+                "content": "esom",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "JEzK8X8AD9hP"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-db81127157a8",
+          "choices": [
+            {
+              "delta": {
+                "content": "eli",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "8EGj5LyQzpZMt"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-db81127157a8",
+          "choices": [
+            {
+              "delta": {
+                "content": "quid",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "wQG19uBuvC7j"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-db81127157a8",
+          "choices": [
+            {
+              "delta": {
+                "content": "\"",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "8Wyenb7E997f9E"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-db81127157a8",
+          "choices": [
+            {
+              "delta": {
+                "content": " is",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "SVXiel7RHA6f3"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-db81127157a8",
+          "choices": [
+            {
+              "delta": {
+                "content": " -",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "ynScunJEjmOWBo"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-db81127157a8",
+          "choices": [
+            {
+              "delta": {
+                "content": "100",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "po2PLlPavc9TN"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-db81127157a8",
+          "choices": [
+            {
+              "delta": {
+                "content": "\u00b0C",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "mt2jiL22pWkH93"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-db81127157a8",
+          "choices": [
+            {
+              "delta": {
+                "content": ".",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "32gJJ61zmjmftOn"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-db81127157a8",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": "stop",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "HszNIiCJ12"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-db81127157a8",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": {
+            "completion_tokens": 17,
+            "prompt_tokens": 188,
+            "total_tokens": 205,
+            "completion_tokens_details": {
+              "accepted_prediction_tokens": 0,
+              "audio_tokens": 0,
+              "reasoning_tokens": 0,
+              "rejected_prediction_tokens": 0
+            },
+            "prompt_tokens_details": {
+              "audio_tokens": 0,
+              "cached_tokens": 0
+            }
+          },
+          "obfuscation": "cAx3IDg7toBDJ"
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/recordings/e2dc09dc546d9b8b99096804fe75fae1f1eb09efe6e4f86c115a78a3db5a59bc.json

@@ -0,0 +1,668 @@
+{
+  "test_id": "tests/integration/responses/test_tool_responses.py::test_response_mcp_tool_approval[openai_client-txt=openai/gpt-4o-True-boiling_point_tool]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "What is the boiling point of myawesomeliquid in Celsius?"
+        },
+        {
+          "role": "assistant",
+          "content": "",
+          "tool_calls": [
+            {
+              "index": 0,
+              "id": "call_bL84OWNnE1s75GJEqGLAK35W",
+              "type": "function",
+              "function": {
+                "name": "get_boiling_point",
+                "arguments": "{\"liquid_name\":\"myawesomeliquid\",\"celsius\":true}"
+              }
+            }
+          ]
+        },
+        {
+          "role": "tool",
+          "tool_call_id": "call_bL84OWNnE1s75GJEqGLAK35W",
+          "content": [
+            {
+              "type": "text",
+              "text": "-100"
+            }
+          ]
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      },
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "greet_everyone",
+            "parameters": {
+              "properties": {
+                "url": {
+                  "title": "Url",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "url"
+              ],
+              "title": "greet_everyoneArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_boiling_point",
+            "description": "\n        Returns the boiling point of a liquid in Celsius or Fahrenheit.\n\n        :param liquid_name: The name of the liquid\n        :param celsius: Whether to return the boiling point in Celsius\n        :return: The boiling point of the liquid in Celcius or Fahrenheit\n        ",
+            "parameters": {
+              "properties": {
+                "liquid_name": {
+                  "title": "Liquid Name",
+                  "type": "string"
+                },
+                "celsius": {
+                  "default": true,
+                  "title": "Celsius",
+                  "type": "boolean"
+                }
+              },
+              "required": [
+                "liquid_name"
+              ],
+              "title": "get_boiling_pointArguments",
+              "type": "object"
+            }
+          }
+        }
+      ]
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e2dc09dc546d",
+          "choices": [
+            {
+              "delta": {
+                "content": "",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "STnb1nbwTsG4JZ"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e2dc09dc546d",
+          "choices": [
+            {
+              "delta": {
+                "content": "The",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "aEUUYMIYjnZpH"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e2dc09dc546d",
+          "choices": [
+            {
+              "delta": {
+                "content": " boiling",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "2QzI8Zau"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e2dc09dc546d",
+          "choices": [
+            {
+              "delta": {
+                "content": " point",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "gZw7vp0bnu"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e2dc09dc546d",
+          "choices": [
+            {
+              "delta": {
+                "content": " of",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "TYru3DcfZVc6B"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e2dc09dc546d",
+          "choices": [
+            {
+              "delta": {
+                "content": " \"",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "h5P3cluszFa21"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e2dc09dc546d",
+          "choices": [
+            {
+              "delta": {
+                "content": "my",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "ggSDGSgtWOR3d9"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e2dc09dc546d",
+          "choices": [
+            {
+              "delta": {
+                "content": "aw",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "lm72CS5Lt7lW76"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e2dc09dc546d",
+          "choices": [
+            {
+              "delta": {
+                "content": "esom",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "fKXRsLB1CG0e"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e2dc09dc546d",
+          "choices": [
+            {
+              "delta": {
+                "content": "eli",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "JxZBNjkfyXquH"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e2dc09dc546d",
+          "choices": [
+            {
+              "delta": {
+                "content": "quid",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "egtKHFRBAqZn"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e2dc09dc546d",
+          "choices": [
+            {
+              "delta": {
+                "content": "\"",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "R7MdHaS5Rj2mMV"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e2dc09dc546d",
+          "choices": [
+            {
+              "delta": {
+                "content": " in",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "LydsYLrAIj6PU"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e2dc09dc546d",
+          "choices": [
+            {
+              "delta": {
+                "content": " Celsius",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "4MmAUDk0"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e2dc09dc546d",
+          "choices": [
+            {
+              "delta": {
+                "content": " is",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Ivlu4M0VfRH8b"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e2dc09dc546d",
+          "choices": [
+            {
+              "delta": {
+                "content": " -",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "OfTmU32oCtMsuo"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e2dc09dc546d",
+          "choices": [
+            {
+              "delta": {
+                "content": "100",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "IUbbHa5oyIPjr"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e2dc09dc546d",
+          "choices": [
+            {
+              "delta": {
+                "content": "\u00b0C",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "llluAF0LBNJIwi"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e2dc09dc546d",
+          "choices": [
+            {
+              "delta": {
+                "content": ".",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "LnUC3LPx43OfUbC"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e2dc09dc546d",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": "stop",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "ULfebGmmMn"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e2dc09dc546d",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": {
+            "completion_tokens": 19,
+            "prompt_tokens": 195,
+            "total_tokens": 214,
+            "completion_tokens_details": {
+              "accepted_prediction_tokens": 0,
+              "audio_tokens": 0,
+              "reasoning_tokens": 0,
+              "rejected_prediction_tokens": 0
+            },
+            "prompt_tokens_details": {
+              "audio_tokens": 0,
+              "cached_tokens": 0
+            }
+          },
+          "obfuscation": "w11BVXjZVXRtg"
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/recordings/e9f1cc3da4297f143b7b2a4b21b34cf2f55727b67c1e1854a106b9d8c7c64b70.json

@@ -0,0 +1,700 @@
+{
+  "test_id": "tests/integration/responses/test_tool_responses.py::test_response_non_streaming_multi_turn_tool_execution[openai_client-txt=openai/gpt-4o-experiment_results_lookup]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "I need to get the results for the 'boiling_point' experiment. First, get the experiment ID for 'boiling_point', then use that ID to get the experiment results. Tell me the boiling point in Celsius."
+        },
+        {
+          "role": "assistant",
+          "content": "",
+          "tool_calls": [
+            {
+              "index": 0,
+              "id": "call_dZwjBxH3aTRhnaS0bJVPqRcz",
+              "type": "function",
+              "function": {
+                "name": "get_experiment_id",
+                "arguments": "{\"experiment_name\":\"boiling_point\"}"
+              }
+            }
+          ]
+        },
+        {
+          "role": "tool",
+          "tool_call_id": "call_dZwjBxH3aTRhnaS0bJVPqRcz",
+          "content": [
+            {
+              "type": "text",
+              "text": "exp_004"
+            }
+          ]
+        },
+        {
+          "role": "assistant",
+          "content": "",
+          "tool_calls": [
+            {
+              "index": 0,
+              "id": "call_skNUKbERbtdoADH834U9OE91",
+              "type": "function",
+              "function": {
+                "name": "get_experiment_results",
+                "arguments": "{\"experiment_id\":\"exp_004\"}"
+              }
+            }
+          ]
+        },
+        {
+          "role": "tool",
+          "tool_call_id": "call_skNUKbERbtdoADH834U9OE91",
+          "content": [
+            {
+              "type": "text",
+              "text": "Boiling Point: 100\u00b0C, Status: Verified"
+            }
+          ]
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      },
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "get_user_id",
+            "description": "\n        Get the user ID for a given username. This ID is needed for other operations.\n\n        :param username: The username to look up\n        :return: The user ID for the username\n        ",
+            "parameters": {
+              "properties": {
+                "username": {
+                  "title": "Username",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "username"
+              ],
+              "title": "get_user_idArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_user_permissions",
+            "description": "\n        Get the permissions for a user ID. Requires a valid user ID from get_user_id.\n\n        :param user_id: The user ID to check permissions for\n        :return: The permissions for the user\n        ",
+            "parameters": {
+              "properties": {
+                "user_id": {
+                  "title": "User Id",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "user_id"
+              ],
+              "title": "get_user_permissionsArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "check_file_access",
+            "description": "\n        Check if a user can access a specific file. Requires a valid user ID.\n\n        :param user_id: The user ID to check access for\n        :param filename: The filename to check access to\n        :return: Whether the user can access the file (yes/no)\n        ",
+            "parameters": {
+              "properties": {
+                "user_id": {
+                  "title": "User Id",
+                  "type": "string"
+                },
+                "filename": {
+                  "title": "Filename",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "user_id",
+                "filename"
+              ],
+              "title": "check_file_accessArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_experiment_id",
+            "description": "\n        Get the experiment ID for a given experiment name. This ID is needed to get results.\n\n        :param experiment_name: The name of the experiment\n        :return: The experiment ID\n        ",
+            "parameters": {
+              "properties": {
+                "experiment_name": {
+                  "title": "Experiment Name",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "experiment_name"
+              ],
+              "title": "get_experiment_idArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_experiment_results",
+            "description": "\n        Get the results for an experiment ID. Requires a valid experiment ID from get_experiment_id.\n\n        :param experiment_id: The experiment ID to get results for\n        :return: The experiment results\n        ",
+            "parameters": {
+              "properties": {
+                "experiment_id": {
+                  "title": "Experiment Id",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "experiment_id"
+              ],
+              "title": "get_experiment_resultsArguments",
+              "type": "object"
+            }
+          }
+        }
+      ]
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e9f1cc3da429",
+          "choices": [
+            {
+              "delta": {
+                "content": "",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "OzNg5nfMI5VouN"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e9f1cc3da429",
+          "choices": [
+            {
+              "delta": {
+                "content": "The",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "EBvjjqFPfytPb"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e9f1cc3da429",
+          "choices": [
+            {
+              "delta": {
+                "content": " boiling",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "HhEiLgKg"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e9f1cc3da429",
+          "choices": [
+            {
+              "delta": {
+                "content": " point",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "hLc2aAgg1D"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e9f1cc3da429",
+          "choices": [
+            {
+              "delta": {
+                "content": " for",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "q3AsmJJ6Rvyt"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e9f1cc3da429",
+          "choices": [
+            {
+              "delta": {
+                "content": " the",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "4QJrcjxcuFLd"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e9f1cc3da429",
+          "choices": [
+            {
+              "delta": {
+                "content": " experiment",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "BQQJ8"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e9f1cc3da429",
+          "choices": [
+            {
+              "delta": {
+                "content": " '",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "nj2SOixVU5KocZ"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e9f1cc3da429",
+          "choices": [
+            {
+              "delta": {
+                "content": "bo",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "ookLm9qkLqQQ3M"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e9f1cc3da429",
+          "choices": [
+            {
+              "delta": {
+                "content": "iling",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "J4axWnSRvQU"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e9f1cc3da429",
+          "choices": [
+            {
+              "delta": {
+                "content": "_point",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "QG6jvQWF8t"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e9f1cc3da429",
+          "choices": [
+            {
+              "delta": {
+                "content": "'",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "veUGdbLd3d8r2yU"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e9f1cc3da429",
+          "choices": [
+            {
+              "delta": {
+                "content": " is",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "ZOCkbhGksYmsF"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e9f1cc3da429",
+          "choices": [
+            {
+              "delta": {
+                "content": " ",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "fbNuaYkAA8gREQ7"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e9f1cc3da429",
+          "choices": [
+            {
+              "delta": {
+                "content": "100",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "3rdZxDq7QoXcl"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e9f1cc3da429",
+          "choices": [
+            {
+              "delta": {
+                "content": "\u00b0C",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "upjHViB9dUBWAd"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e9f1cc3da429",
+          "choices": [
+            {
+              "delta": {
+                "content": ".",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "hBZNqRjyLGCIMjg"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e9f1cc3da429",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": "stop",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "PrtgvDwRZp"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-e9f1cc3da429",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": {
+            "completion_tokens": 17,
+            "prompt_tokens": 490,
+            "total_tokens": 507,
+            "completion_tokens_details": {
+              "accepted_prediction_tokens": 0,
+              "audio_tokens": 0,
+              "reasoning_tokens": 0,
+              "rejected_prediction_tokens": 0
+            },
+            "prompt_tokens_details": {
+              "audio_tokens": 0,
+              "cached_tokens": 0
+            }
+          },
+          "obfuscation": "euYYBnLE4Mj0Z"
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/recordings/ed89b57fec937fa8602b4911a21a9a1a9488fb2347bf73d6e3bc2203a9a47a61.json

@@ -0,0 +1,641 @@
+{
+  "test_id": "tests/integration/responses/test_tool_responses.py::test_response_sequential_mcp_tool[openai_client-txt=openai/gpt-4o-boiling_point_tool]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "What is the boiling point of myawesomeliquid in Celsius?"
+        },
+        {
+          "role": "assistant",
+          "content": "",
+          "tool_calls": [
+            {
+              "index": 0,
+              "id": "call_b5k2yeqIi5ucElnnrVPyYU4x",
+              "type": "function",
+              "function": {
+                "name": "get_boiling_point",
+                "arguments": "{\"liquid_name\":\"myawesomeliquid\",\"celsius\":true}"
+              }
+            }
+          ]
+        },
+        {
+          "role": "tool",
+          "tool_call_id": "call_b5k2yeqIi5ucElnnrVPyYU4x",
+          "content": [
+            {
+              "type": "text",
+              "text": "-100"
+            }
+          ]
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      },
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "greet_everyone",
+            "parameters": {
+              "properties": {
+                "url": {
+                  "title": "Url",
+                  "type": "string"
+                }
+              },
+              "required": [
+                "url"
+              ],
+              "title": "greet_everyoneArguments",
+              "type": "object"
+            }
+          }
+        },
+        {
+          "type": "function",
+          "function": {
+            "name": "get_boiling_point",
+            "description": "\n        Returns the boiling point of a liquid in Celsius or Fahrenheit.\n\n        :param liquid_name: The name of the liquid\n        :param celsius: Whether to return the boiling point in Celsius\n        :return: The boiling point of the liquid in Celcius or Fahrenheit\n        ",
+            "parameters": {
+              "properties": {
+                "liquid_name": {
+                  "title": "Liquid Name",
+                  "type": "string"
+                },
+                "celsius": {
+                  "default": true,
+                  "title": "Celsius",
+                  "type": "boolean"
+                }
+              },
+              "required": [
+                "liquid_name"
+              ],
+              "title": "get_boiling_pointArguments",
+              "type": "object"
+            }
+          }
+        }
+      ]
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-ed89b57fec93",
+          "choices": [
+            {
+              "delta": {
+                "content": "",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "WGXCgkwfwMDUCG"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-ed89b57fec93",
+          "choices": [
+            {
+              "delta": {
+                "content": "The",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "pkdvw6gGNrtXN"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-ed89b57fec93",
+          "choices": [
+            {
+              "delta": {
+                "content": " boiling",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "RO5YJeZc"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-ed89b57fec93",
+          "choices": [
+            {
+              "delta": {
+                "content": " point",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "riZZHSDEz0"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-ed89b57fec93",
+          "choices": [
+            {
+              "delta": {
+                "content": " of",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "1zjk8zIdt2Y2b"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-ed89b57fec93",
+          "choices": [
+            {
+              "delta": {
+                "content": " \"",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "XGHv0dlif7IrC"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-ed89b57fec93",
+          "choices": [
+            {
+              "delta": {
+                "content": "my",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Ii2KeTyV3U0uiU"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-ed89b57fec93",
+          "choices": [
+            {
+              "delta": {
+                "content": "aw",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "3OyYvSytdOYhpT"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-ed89b57fec93",
+          "choices": [
+            {
+              "delta": {
+                "content": "esom",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "zCnXbjW4JE6l"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-ed89b57fec93",
+          "choices": [
+            {
+              "delta": {
+                "content": "eli",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "0bwcz2K91q7EO"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-ed89b57fec93",
+          "choices": [
+            {
+              "delta": {
+                "content": "quid",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Um0jFlJegpXI"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-ed89b57fec93",
+          "choices": [
+            {
+              "delta": {
+                "content": "\"",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "4OllZlS2JmoD3l"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-ed89b57fec93",
+          "choices": [
+            {
+              "delta": {
+                "content": " is",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "x4jApO80AyXpX"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-ed89b57fec93",
+          "choices": [
+            {
+              "delta": {
+                "content": " -",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "wq0D3Wzc1l3h6S"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-ed89b57fec93",
+          "choices": [
+            {
+              "delta": {
+                "content": "100",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Dn78V58iZ9wKK"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-ed89b57fec93",
+          "choices": [
+            {
+              "delta": {
+                "content": " degrees",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "fjHDBTqT"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-ed89b57fec93",
+          "choices": [
+            {
+              "delta": {
+                "content": " Celsius",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "Cnp6KULL"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-ed89b57fec93",
+          "choices": [
+            {
+              "delta": {
+                "content": ".",
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "grbygHexDT4JwGx"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-ed89b57fec93",
+          "choices": [
+            {
+              "delta": {
+                "content": null,
+                "function_call": null,
+                "refusal": null,
+                "role": null,
+                "tool_calls": null
+              },
+              "finish_reason": "stop",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": null,
+          "obfuscation": "upSRpiQQKE"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-ed89b57fec93",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": {
+            "completion_tokens": 18,
+            "prompt_tokens": 195,
+            "total_tokens": 213,
+            "completion_tokens_details": {
+              "accepted_prediction_tokens": 0,
+              "audio_tokens": 0,
+              "reasoning_tokens": 0,
+              "rejected_prediction_tokens": 0
+            },
+            "prompt_tokens_details": {
+              "audio_tokens": 0,
+              "cached_tokens": 0
+            }
+          },
+          "obfuscation": "psE6Es6zZ2Kz4"
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/test_basic_responses.py

@@ -13,8 +13,8 @@ from .streaming_assertions import StreamingValidator
 
 
 @pytest.mark.parametrize("case", basic_test_cases)
-def test_response_non_streaming_basic(compat_client, text_model_id, case):
-    response = compat_client.responses.create(
+def test_response_non_streaming_basic(responses_client, text_model_id, case):
+    response = responses_client.responses.create(
         model=text_model_id,
         input=case.input,
         stream=False,
@@ -31,10 +31,10 @@ def test_response_non_streaming_basic(compat_client, text_model_id, case):
         "Total tokens should equal input + output tokens"
     )
 
-    retrieved_response = compat_client.responses.retrieve(response_id=response.id)
+    retrieved_response = responses_client.responses.retrieve(response_id=response.id)
     assert retrieved_response.output_text == response.output_text
 
-    next_response = compat_client.responses.create(
+    next_response = responses_client.responses.create(
         model=text_model_id,
         input="Repeat your previous response in all caps.",
         previous_response_id=response.id,
@@ -44,8 +44,8 @@ def test_response_non_streaming_basic(compat_client, text_model_id, case):
 
 
 @pytest.mark.parametrize("case", basic_test_cases)
-def test_response_streaming_basic(compat_client, text_model_id, case):
-    response = compat_client.responses.create(
+def test_response_streaming_basic(responses_client, text_model_id, case):
+    response = responses_client.responses.create(
         model=text_model_id,
         input=case.input,
         stream=True,
@@ -98,15 +98,15 @@ def test_response_streaming_basic(compat_client, text_model_id, case):
     validator.assert_response_consistency()
 
     # Verify stored response matches streamed response
-    retrieved_response = compat_client.responses.retrieve(response_id=response_id)
+    retrieved_response = responses_client.responses.retrieve(response_id=response_id)
     final_event = events[-1]
     assert retrieved_response.output_text == final_event.response.output_text
 
 
 @pytest.mark.parametrize("case", basic_test_cases)
-def test_response_streaming_incremental_content(compat_client, text_model_id, case):
+def test_response_streaming_incremental_content(responses_client, text_model_id, case):
     """Test that streaming actually delivers content incrementally, not just at the end."""
-    response = compat_client.responses.create(
+    response = responses_client.responses.create(
         model=text_model_id,
         input=case.input,
         stream=True,
@@ -170,10 +170,10 @@ def test_response_streaming_incremental_content(compat_client, text_model_id, ca
 
 
 @pytest.mark.parametrize("case", multi_turn_test_cases)
-def test_response_non_streaming_multi_turn(compat_client, text_model_id, case):
+def test_response_non_streaming_multi_turn(responses_client, text_model_id, case):
     previous_response_id = None
     for turn_input, turn_expected in case.turns:
-        response = compat_client.responses.create(
+        response = responses_client.responses.create(
             model=text_model_id,
             input=turn_input,
             previous_response_id=previous_response_id,
@@ -184,8 +184,8 @@ def test_response_non_streaming_multi_turn(compat_client, text_model_id, case):
 
 
 @pytest.mark.parametrize("case", image_test_cases)
-def test_response_non_streaming_image(compat_client, text_model_id, case):
-    response = compat_client.responses.create(
+def test_response_non_streaming_image(responses_client, text_model_id, case):
+    response = responses_client.responses.create(
         model=text_model_id,
         input=case.input,
         stream=False,
@@ -195,10 +195,10 @@ def test_response_non_streaming_image(compat_client, text_model_id, case):
 
 
 @pytest.mark.parametrize("case", multi_turn_image_test_cases)
-def test_response_non_streaming_multi_turn_image(compat_client, text_model_id, case):
+def test_response_non_streaming_multi_turn_image(responses_client, text_model_id, case):
     previous_response_id = None
     for turn_input, turn_expected in case.turns:
-        response = compat_client.responses.create(
+        response = responses_client.responses.create(
             model=text_model_id,
             input=turn_input,
             previous_response_id=previous_response_id,

tests/integration/responses/test_conversation_responses.py

@@ -65,8 +65,14 @@ class TestConversationResponses:
         conversation_items = openai_client.conversations.items.list(conversation.id)
         assert len(conversation_items.data) >= 4  # 2 user + 2 assistant messages
 
+    @pytest.mark.timeout(60, method="thread")
     def test_conversation_context_loading(self, openai_client, text_model_id):
-        """Test that conversation context is properly loaded for responses."""
+        """Test that conversation context is properly loaded for responses.
+
+        Note: 60s timeout added due to CI-specific deadlock in pytest/OpenAI client/httpx
+        after running 25+ tests. Hangs before first HTTP request is made. Works fine locally.
+        Investigation needed: connection pool exhaustion or event loop state issue.
+        """
         conversation = openai_client.conversations.create(
             items=[
                 {"type": "message", "role": "user", "content": "My name is Alice. I like to eat apples."},
@@ -82,6 +88,7 @@ class TestConversationResponses:
 
         assert "apple" in response.output_text.lower()
 
+    @pytest.mark.timeout(60, method="thread")
     def test_conversation_error_handling(self, openai_client, text_model_id):
         """Test error handling for invalid and nonexistent conversations."""
         # Invalid conversation ID format
@@ -125,18 +132,18 @@ class TestConversationResponses:
         assert len(response.output_text.strip()) > 0
 
     # this is not ready yet
-    # def test_conversation_compat_client(self, compat_client, text_model_id):
+    # def test_conversation_compat_client(self, responses_client, text_model_id):
     #     """Test conversation parameter works with compatibility client."""
-    #     if not hasattr(compat_client, "conversations"):
-    #         pytest.skip("compat_client does not support conversations API")
+    #     if not hasattr(responses_client, "conversations"):
+    #         pytest.skip("responses_client does not support conversations API")
     #
-    #     conversation = compat_client.conversations.create()
-    #     response = compat_client.responses.create(
+    #     conversation = responses_client.conversations.create()
+    #     response = responses_client.responses.create(
     #         model=text_model_id, input="Tell me a joke", conversation=conversation.id
     #     )
     #
     #     assert response is not None
     #     assert len(response.output_text.strip()) > 0
     #
-    #     conversation_items = compat_client.conversations.items.list(conversation.id)
+    #     conversation_items = responses_client.conversations.items.list(conversation.id)
     #     assert len(conversation_items.data) >= 2

tests/integration/responses/test_file_search.py

@@ -9,8 +9,6 @@ import time
 
 import pytest
 
-from llama_stack.core.library_client import LlamaStackAsLibraryClient
-
 from .helpers import new_vector_store, upload_file
 
 
@@ -28,12 +26,9 @@ from .helpers import new_vector_store, upload_file
         },
     ],
 )
-def test_response_text_format(compat_client, text_model_id, text_format):
-    if isinstance(compat_client, LlamaStackAsLibraryClient):
-        pytest.skip("Responses API text format is not yet supported in library client.")
-
+def test_response_text_format(responses_client, text_model_id, text_format):
     stream = False
-    response = compat_client.responses.create(
+    response = responses_client.responses.create(
         model=text_model_id,
         input="What is the capital of France?",
         stream=stream,
@@ -47,13 +42,10 @@ def test_response_text_format(compat_client, text_model_id, text_format):
 
 
 @pytest.fixture
-def vector_store_with_filtered_files(compat_client, embedding_model_id, embedding_dimension, tmp_path_factory):
+def vector_store_with_filtered_files(responses_client, embedding_model_id, embedding_dimension, tmp_path_factory):
     # """Create a vector store with multiple files that have different attributes for filtering tests."""
-    if isinstance(compat_client, LlamaStackAsLibraryClient):
-        pytest.skip("upload_file() is not yet supported in library client somehow?")
-
     vector_store = new_vector_store(
-        compat_client, "test_vector_store_with_filters", embedding_model_id, embedding_dimension
+        responses_client, "test_vector_store_with_filters", embedding_model_id, embedding_dimension
     )
     tmp_path = tmp_path_factory.mktemp("filter_test_files")
 
@@ -104,11 +96,11 @@ def vector_store_with_filtered_files(compat_client, embedding_model_id, embeddin
         file_path.write_text(file_data["content"])
 
         # Upload file
-        file_response = upload_file(compat_client, file_data["name"], str(file_path))
+        file_response = upload_file(responses_client, file_data["name"], str(file_path))
         file_ids.append(file_response.id)
 
         # Attach file to vector store with attributes
-        file_attach_response = compat_client.vector_stores.files.create(
+        file_attach_response = responses_client.vector_stores.files.create(
             vector_store_id=vector_store.id,
             file_id=file_response.id,
             attributes=file_data["attributes"],
@@ -117,7 +109,7 @@ def vector_store_with_filtered_files(compat_client, embedding_model_id, embeddin
         # Wait for attachment
         while file_attach_response.status == "in_progress":
             time.sleep(0.1)
-            file_attach_response = compat_client.vector_stores.files.retrieve(
+            file_attach_response = responses_client.vector_stores.files.retrieve(
                 vector_store_id=vector_store.id,
                 file_id=file_response.id,
             )
@@ -127,17 +119,17 @@ def vector_store_with_filtered_files(compat_client, embedding_model_id, embeddin
 
     # Cleanup: delete vector store and files
     try:
-        compat_client.vector_stores.delete(vector_store_id=vector_store.id)
+        responses_client.vector_stores.delete(vector_store_id=vector_store.id)
         for file_id in file_ids:
             try:
-                compat_client.files.delete(file_id=file_id)
+                responses_client.files.delete(file_id=file_id)
             except Exception:
                 pass  # File might already be deleted
     except Exception:
         pass  # Best effort cleanup
 
 
-def test_response_file_search_filter_by_region(compat_client, text_model_id, vector_store_with_filtered_files):
+def test_response_file_search_filter_by_region(responses_client, text_model_id, vector_store_with_filtered_files):
     """Test file search with region equality filter."""
     tools = [
         {
@@ -147,7 +139,7 @@ def test_response_file_search_filter_by_region(compat_client, text_model_id, vec
         }
     ]
 
-    response = compat_client.responses.create(
+    response = responses_client.responses.create(
         model=text_model_id,
         input="What are the updates from the US region?",
         tools=tools,
@@ -168,7 +160,7 @@ def test_response_file_search_filter_by_region(compat_client, text_model_id, vec
         assert "asia" not in result.text.lower()
 
 
-def test_response_file_search_filter_by_category(compat_client, text_model_id, vector_store_with_filtered_files):
+def test_response_file_search_filter_by_category(responses_client, text_model_id, vector_store_with_filtered_files):
     """Test file search with category equality filter."""
     tools = [
         {
@@ -178,7 +170,7 @@ def test_response_file_search_filter_by_category(compat_client, text_model_id, v
         }
     ]
 
-    response = compat_client.responses.create(
+    response = responses_client.responses.create(
         model=text_model_id,
         input="Show me all marketing reports",
         tools=tools,
@@ -198,7 +190,7 @@ def test_response_file_search_filter_by_category(compat_client, text_model_id, v
         assert "revenue figures" not in result.text.lower()
 
 
-def test_response_file_search_filter_by_date_range(compat_client, text_model_id, vector_store_with_filtered_files):
+def test_response_file_search_filter_by_date_range(responses_client, text_model_id, vector_store_with_filtered_files):
     """Test file search with date range filter using compound AND."""
     tools = [
         {
@@ -222,7 +214,7 @@ def test_response_file_search_filter_by_date_range(compat_client, text_model_id,
         }
     ]
 
-    response = compat_client.responses.create(
+    response = responses_client.responses.create(
         model=text_model_id,
         input="What happened in Q1 2023?",
         tools=tools,
@@ -241,7 +233,7 @@ def test_response_file_search_filter_by_date_range(compat_client, text_model_id,
         assert "q3" not in result.text.lower()
 
 
-def test_response_file_search_filter_compound_and(compat_client, text_model_id, vector_store_with_filtered_files):
+def test_response_file_search_filter_compound_and(responses_client, text_model_id, vector_store_with_filtered_files):
     """Test file search with compound AND filter (region AND category)."""
     tools = [
         {
@@ -257,7 +249,7 @@ def test_response_file_search_filter_compound_and(compat_client, text_model_id,
         }
     ]
 
-    response = compat_client.responses.create(
+    response = responses_client.responses.create(
         model=text_model_id,
         input="What are the engineering updates from the US?",
         tools=tools,
@@ -277,7 +269,7 @@ def test_response_file_search_filter_compound_and(compat_client, text_model_id,
         assert "promotional" not in result.text.lower() and "revenue" not in result.text.lower()
 
 
-def test_response_file_search_filter_compound_or(compat_client, text_model_id, vector_store_with_filtered_files):
+def test_response_file_search_filter_compound_or(responses_client, text_model_id, vector_store_with_filtered_files):
     """Test file search with compound OR filter (marketing OR sales)."""
     tools = [
         {
@@ -293,7 +285,7 @@ def test_response_file_search_filter_compound_or(compat_client, text_model_id, v
         }
     ]
 
-    response = compat_client.responses.create(
+    response = responses_client.responses.create(
         model=text_model_id,
         input="Show me marketing and sales documents",
         tools=tools,
@@ -320,7 +312,7 @@ def test_response_file_search_filter_compound_or(compat_client, text_model_id, v
     assert categories_found.issubset({"marketing", "sales"}), f"Found unexpected categories: {categories_found}"
 
 
-def test_response_file_search_streaming_events(compat_client, text_model_id, vector_store_with_filtered_files):
+def test_response_file_search_streaming_events(responses_client, text_model_id, vector_store_with_filtered_files):
     """Test that file search emits proper streaming events (in_progress, searching, completed)."""
     tools = [
         {
@@ -329,7 +321,7 @@ def test_response_file_search_streaming_events(compat_client, text_model_id, vec
         }
     ]
 
-    stream = compat_client.responses.create(
+    stream = responses_client.responses.create(
         model=text_model_id,
         input="What are the marketing updates?",
         tools=tools,

tests/integration/responses/test_mcp_authentication.py

@@ -0,0 +1,105 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+
+import pytest
+
+from tests.common.mcp import make_mcp_server
+
+from .helpers import setup_mcp_tools
+
+# MCP authentication tests with recordings
+# Tests for bearer token authorization support in MCP tool configurations
+
+
+def test_mcp_authorization_bearer(responses_client, text_model_id):
+    """Test that bearer authorization is correctly applied to MCP requests."""
+    test_token = "test-bearer-token-789"
+    with make_mcp_server(required_auth_token=test_token) as mcp_server_info:
+        tools = setup_mcp_tools(
+            [
+                {
+                    "type": "mcp",
+                    "server_label": "auth-mcp",
+                    "server_url": "<FILLED_BY_TEST_RUNNER>",
+                    "authorization": test_token,  # Just the token, not "Bearer <token>"
+                }
+            ],
+            mcp_server_info,
+        )
+
+        # Create response - authorization should be applied
+        response = responses_client.responses.create(
+            model=text_model_id,
+            input="What is the boiling point of myawesomeliquid?",
+            tools=tools,
+            stream=False,
+        )
+
+        # Verify list_tools succeeded (requires auth)
+        assert len(response.output) >= 3
+        assert response.output[0].type == "mcp_list_tools"
+        assert len(response.output[0].tools) == 2
+
+        # Verify tool invocation succeeded (requires auth)
+        assert response.output[1].type == "mcp_call"
+        assert response.output[1].error is None
+
+
+def test_mcp_authorization_error_when_header_provided(responses_client, text_model_id):
+    """Test that providing Authorization in headers raises a security error."""
+    test_token = "test-token-123"
+    with make_mcp_server(required_auth_token=test_token) as mcp_server_info:
+        tools = setup_mcp_tools(
+            [
+                {
+                    "type": "mcp",
+                    "server_label": "header-auth-mcp",
+                    "server_url": "<FILLED_BY_TEST_RUNNER>",
+                    "headers": {"Authorization": f"Bearer {test_token}"},  # Security risk - should be rejected
+                }
+            ],
+            mcp_server_info,
+        )
+
+        # Create response - should raise BadRequestError for security reasons
+        with pytest.raises((ValueError, Exception), match="Authorization header cannot be passed via 'headers'"):
+            responses_client.responses.create(
+                model=text_model_id,
+                input="What is the boiling point of myawesomeliquid?",
+                tools=tools,
+                stream=False,
+            )
+
+
+def test_mcp_authorization_backward_compatibility(responses_client, text_model_id):
+    """Test that MCP tools work without authorization (backward compatibility)."""
+    # No authorization required
+    with make_mcp_server(required_auth_token=None) as mcp_server_info:
+        tools = setup_mcp_tools(
+            [
+                {
+                    "type": "mcp",
+                    "server_label": "noauth-mcp",
+                    "server_url": "<FILLED_BY_TEST_RUNNER>",
+                }
+            ],
+            mcp_server_info,
+        )
+
+        # Create response without authorization
+        response = responses_client.responses.create(
+            model=text_model_id,
+            input="What is the boiling point of myawesomeliquid?",
+            tools=tools,
+            stream=False,
+        )
+
+        # Verify operations succeeded without auth
+        assert len(response.output) >= 3
+        assert response.output[0].type == "mcp_list_tools"
+        assert response.output[1].type == "mcp_call"
+        assert response.output[1].error is None

tests/integration/responses/test_tool_responses.py

@@ -9,6 +9,7 @@ import logging  # allow-direct-logging
 import os
 
 import httpx
+import llama_stack_client
 import openai
 import pytest
 
@@ -29,8 +30,8 @@ from .streaming_assertions import StreamingValidator
 
 
 @pytest.mark.parametrize("case", web_search_test_cases)
-def test_response_non_streaming_web_search(compat_client, text_model_id, case):
-    response = compat_client.responses.create(
+def test_response_non_streaming_web_search(responses_client, text_model_id, case):
+    response = responses_client.responses.create(
         model=text_model_id,
         input=case.input,
         tools=case.tools,
@@ -48,12 +49,9 @@ def test_response_non_streaming_web_search(compat_client, text_model_id, case):
 
 @pytest.mark.parametrize("case", file_search_test_cases)
 def test_response_non_streaming_file_search(
-    compat_client, text_model_id, embedding_model_id, embedding_dimension, tmp_path, case
+    responses_client, text_model_id, embedding_model_id, embedding_dimension, tmp_path, case
 ):
-    if isinstance(compat_client, LlamaStackAsLibraryClient):
-        pytest.skip("Responses API file search is not yet supported in library client.")
-
-    vector_store = new_vector_store(compat_client, "test_vector_store", embedding_model_id, embedding_dimension)
+    vector_store = new_vector_store(responses_client, "test_vector_store", embedding_model_id, embedding_dimension)
 
     if case.file_content:
         file_name = "test_response_non_streaming_file_search.txt"
@@ -65,16 +63,16 @@ def test_response_non_streaming_file_search(
     else:
         raise ValueError("No file content or path provided for case")
 
-    file_response = upload_file(compat_client, file_name, file_path)
+    file_response = upload_file(responses_client, file_name, file_path)
 
     # Attach our file to the vector store
-    compat_client.vector_stores.files.create(
+    responses_client.vector_stores.files.create(
         vector_store_id=vector_store.id,
         file_id=file_response.id,
     )
 
     # Wait for the file to be attached
-    wait_for_file_attachment(compat_client, vector_store.id, file_response.id)
+    wait_for_file_attachment(responses_client, vector_store.id, file_response.id)
 
     # Update our tools with the right vector store id
     tools = case.tools
@@ -83,7 +81,7 @@ def test_response_non_streaming_file_search(
             tool["vector_store_ids"] = [vector_store.id]
 
     # Create the response request, which should query our vector store
-    response = compat_client.responses.create(
+    response = responses_client.responses.create(
         model=text_model_id,
         input=case.input,
         tools=tools,
@@ -105,15 +103,12 @@ def test_response_non_streaming_file_search(
 
 
 def test_response_non_streaming_file_search_empty_vector_store(
-    compat_client, text_model_id, embedding_model_id, embedding_dimension
+    responses_client, text_model_id, embedding_model_id, embedding_dimension
 ):
-    if isinstance(compat_client, LlamaStackAsLibraryClient):
-        pytest.skip("Responses API file search is not yet supported in library client.")
-
-    vector_store = new_vector_store(compat_client, "test_vector_store", embedding_model_id, embedding_dimension)
+    vector_store = new_vector_store(responses_client, "test_vector_store", embedding_model_id, embedding_dimension)
 
     # Create the response request, which should query our vector store
-    response = compat_client.responses.create(
+    response = responses_client.responses.create(
         model=text_model_id,
         input="How many experts does the Llama 4 Maverick model have?",
         tools=[{"type": "file_search", "vector_store_ids": [vector_store.id]}],
@@ -133,13 +128,10 @@ def test_response_non_streaming_file_search_empty_vector_store(
 
 
 def test_response_sequential_file_search(
-    compat_client, text_model_id, embedding_model_id, embedding_dimension, tmp_path
+    responses_client, text_model_id, embedding_model_id, embedding_dimension, tmp_path
 ):
     """Test file search with sequential responses using previous_response_id."""
-    if isinstance(compat_client, LlamaStackAsLibraryClient):
-        pytest.skip("Responses API file search is not yet supported in library client.")
-
-    vector_store = new_vector_store(compat_client, "test_vector_store", embedding_model_id, embedding_dimension)
+    vector_store = new_vector_store(responses_client, "test_vector_store", embedding_model_id, embedding_dimension)
 
     # Create a test file with content
     file_content = "The Llama 4 Maverick model has 128 experts in its mixture of experts architecture."
@@ -147,21 +139,21 @@ def test_response_sequential_file_search(
     file_path = tmp_path / file_name
     file_path.write_text(file_content)
 
-    file_response = upload_file(compat_client, file_name, file_path)
+    file_response = upload_file(responses_client, file_name, file_path)
 
     # Attach the file to the vector store
-    compat_client.vector_stores.files.create(
+    responses_client.vector_stores.files.create(
         vector_store_id=vector_store.id,
         file_id=file_response.id,
     )
 
     # Wait for the file to be attached
-    wait_for_file_attachment(compat_client, vector_store.id, file_response.id)
+    wait_for_file_attachment(responses_client, vector_store.id, file_response.id)
 
     tools = [{"type": "file_search", "vector_store_ids": [vector_store.id]}]
 
     # First response request with file search
-    response = compat_client.responses.create(
+    response = responses_client.responses.create(
         model=text_model_id,
         input="How many experts does the Llama 4 Maverick model have?",
         tools=tools,
@@ -178,7 +170,7 @@ def test_response_sequential_file_search(
     assert "128" in response.output_text or "experts" in response.output_text.lower()
 
     # Second response request using previous_response_id
-    response2 = compat_client.responses.create(
+    response2 = responses_client.responses.create(
         model=text_model_id,
         input="Can you tell me more about the architecture?",
         tools=tools,
@@ -199,14 +191,11 @@ def test_response_sequential_file_search(
 
 
 @pytest.mark.parametrize("case", mcp_tool_test_cases)
-def test_response_non_streaming_mcp_tool(compat_client, text_model_id, case, caplog):
-    if not isinstance(compat_client, LlamaStackAsLibraryClient):
-        pytest.skip("in-process MCP server is only supported in library client")
-
+def test_response_non_streaming_mcp_tool(responses_client, text_model_id, case, caplog):
     with make_mcp_server() as mcp_server_info:
         tools = setup_mcp_tools(case.tools, mcp_server_info)
 
-        response = compat_client.responses.create(
+        response = responses_client.responses.create(
             model=text_model_id,
             input=case.input,
             tools=tools,
@@ -243,15 +232,15 @@ def test_response_non_streaming_mcp_tool(compat_client, text_model_id, case, cap
 
         exc_type = (
             AuthenticationRequiredError
-            if isinstance(compat_client, LlamaStackAsLibraryClient)
-            else (httpx.HTTPStatusError, openai.AuthenticationError)
+            if isinstance(responses_client, LlamaStackAsLibraryClient)
+            else (httpx.HTTPStatusError, openai.AuthenticationError, llama_stack_client.AuthenticationError)
         )
         # Suppress expected auth error logs only for the failing auth attempt
         with caplog.at_level(
             logging.CRITICAL, logger="llama_stack.providers.inline.agents.meta_reference.responses.streaming"
         ):
             with pytest.raises(exc_type):
-                compat_client.responses.create(
+                responses_client.responses.create(
                     model=text_model_id,
                     input=case.input,
                     tools=tools,
@@ -260,9 +249,9 @@ def test_response_non_streaming_mcp_tool(compat_client, text_model_id, case, cap
 
         for tool in tools:
             if tool["type"] == "mcp":
-                tool["headers"] = {"Authorization": "Bearer test-token"}
+                tool["authorization"] = "test-token"
 
-        response = compat_client.responses.create(
+        response = responses_client.responses.create(
             model=text_model_id,
             input=case.input,
             tools=tools,
@@ -272,14 +261,11 @@ def test_response_non_streaming_mcp_tool(compat_client, text_model_id, case, cap
 
 
 @pytest.mark.parametrize("case", mcp_tool_test_cases)
-def test_response_sequential_mcp_tool(compat_client, text_model_id, case):
-    if not isinstance(compat_client, LlamaStackAsLibraryClient):
-        pytest.skip("in-process MCP server is only supported in library client")
-
+def test_response_sequential_mcp_tool(responses_client, text_model_id, case):
     with make_mcp_server() as mcp_server_info:
         tools = setup_mcp_tools(case.tools, mcp_server_info)
 
-        response = compat_client.responses.create(
+        response = responses_client.responses.create(
             model=text_model_id,
             input=case.input,
             tools=tools,
@@ -311,7 +297,7 @@ def test_response_sequential_mcp_tool(compat_client, text_model_id, case):
         text_content = message.content[0].text
         assert "boiling point" in text_content.lower()
 
-        response2 = compat_client.responses.create(
+        response2 = responses_client.responses.create(
             model=text_model_id, input=case.input, tools=tools, stream=False, previous_response_id=response.id
         )
 
@@ -323,16 +309,13 @@ def test_response_sequential_mcp_tool(compat_client, text_model_id, case):
 
 @pytest.mark.parametrize("case", mcp_tool_test_cases)
 @pytest.mark.parametrize("approve", [True, False])
-def test_response_mcp_tool_approval(compat_client, text_model_id, case, approve):
-    if not isinstance(compat_client, LlamaStackAsLibraryClient):
-        pytest.skip("in-process MCP server is only supported in library client")
-
+def test_response_mcp_tool_approval(responses_client, text_model_id, case, approve):
     with make_mcp_server() as mcp_server_info:
         tools = setup_mcp_tools(case.tools, mcp_server_info)
         for tool in tools:
             tool["require_approval"] = "always"
 
-        response = compat_client.responses.create(
+        response = responses_client.responses.create(
             model=text_model_id,
             input=case.input,
             tools=tools,
@@ -352,13 +335,13 @@ def test_response_mcp_tool_approval(compat_client, text_model_id, case, approve)
         approval_request = response.output[1]
         assert approval_request.type == "mcp_approval_request"
         assert approval_request.name == "get_boiling_point"
-        assert json.loads(approval_request.arguments) == {
-            "liquid_name": "myawesomeliquid",
-            "celsius": True,
-        }
+        args = json.loads(approval_request.arguments)
+        assert args["liquid_name"] == "myawesomeliquid"
+        # celsius has a default value of True, so it may be omitted or explicitly set
+        assert args.get("celsius", True) is True
 
         # send approval response
-        response = compat_client.responses.create(
+        response = responses_client.responses.create(
             previous_response_id=response.id,
             model=text_model_id,
             input=[{"type": "mcp_approval_response", "approval_request_id": approval_request.id, "approve": approve}],
@@ -398,8 +381,8 @@ def test_response_mcp_tool_approval(compat_client, text_model_id, case, approve)
 
 
 @pytest.mark.parametrize("case", custom_tool_test_cases)
-def test_response_non_streaming_custom_tool(compat_client, text_model_id, case):
-    response = compat_client.responses.create(
+def test_response_non_streaming_custom_tool(responses_client, text_model_id, case):
+    response = responses_client.responses.create(
         model=text_model_id,
         input=case.input,
         tools=case.tools,
@@ -412,8 +395,8 @@ def test_response_non_streaming_custom_tool(compat_client, text_model_id, case):
 
 
 @pytest.mark.parametrize("case", custom_tool_test_cases)
-def test_response_function_call_ordering_1(compat_client, text_model_id, case):
-    response = compat_client.responses.create(
+def test_response_function_call_ordering_1(responses_client, text_model_id, case):
+    response = responses_client.responses.create(
         model=text_model_id,
         input=case.input,
         tools=case.tools,
@@ -437,13 +420,13 @@ def test_response_function_call_ordering_1(compat_client, text_model_id, case):
             "call_id": response.output[0].call_id,
         }
     )
-    response = compat_client.responses.create(
+    response = responses_client.responses.create(
         model=text_model_id, input=inputs, tools=case.tools, stream=False, previous_response_id=response.id
     )
     assert len(response.output) == 1
 
 
-def test_response_function_call_ordering_2(compat_client, text_model_id):
+def test_response_function_call_ordering_2(responses_client, text_model_id):
     tools = [
         {
             "type": "function",
@@ -468,7 +451,7 @@ def test_response_function_call_ordering_2(compat_client, text_model_id):
             "content": "Is the weather better in San Francisco or Los Angeles?",
         }
     ]
-    response = compat_client.responses.create(
+    response = responses_client.responses.create(
         model=text_model_id,
         input=inputs,
         tools=tools,
@@ -489,7 +472,7 @@ def test_response_function_call_ordering_2(compat_client, text_model_id):
                     "call_id": output.call_id,
                 }
             )
-    response = compat_client.responses.create(
+    response = responses_client.responses.create(
         model=text_model_id,
         input=inputs,
         tools=tools,
@@ -500,15 +483,12 @@ def test_response_function_call_ordering_2(compat_client, text_model_id):
 
 
 @pytest.mark.parametrize("case", multi_turn_tool_execution_test_cases)
-def test_response_non_streaming_multi_turn_tool_execution(compat_client, text_model_id, case):
+def test_response_non_streaming_multi_turn_tool_execution(responses_client, text_model_id, case):
     """Test multi-turn tool execution where multiple MCP tool calls are performed in sequence."""
-    if not isinstance(compat_client, LlamaStackAsLibraryClient):
-        pytest.skip("in-process MCP server is only supported in library client")
-
     with make_mcp_server(tools=dependency_tools()) as mcp_server_info:
         tools = setup_mcp_tools(case.tools, mcp_server_info)
 
-        response = compat_client.responses.create(
+        response = responses_client.responses.create(
             input=case.input,
             model=text_model_id,
             tools=tools,
@@ -550,15 +530,12 @@ def test_response_non_streaming_multi_turn_tool_execution(compat_client, text_mo
 
 
 @pytest.mark.parametrize("case", multi_turn_tool_execution_streaming_test_cases)
-def test_response_streaming_multi_turn_tool_execution(compat_client, text_model_id, case):
+def test_response_streaming_multi_turn_tool_execution(responses_client, text_model_id, case):
     """Test streaming multi-turn tool execution where multiple MCP tool calls are performed in sequence."""
-    if not isinstance(compat_client, LlamaStackAsLibraryClient):
-        pytest.skip("in-process MCP server is only supported in library client")
-
     with make_mcp_server(tools=dependency_tools()) as mcp_server_info:
         tools = setup_mcp_tools(case.tools, mcp_server_info)
 
-        stream = compat_client.responses.create(
+        stream = responses_client.responses.create(
             input=case.input,
             model=text_model_id,
             tools=tools,

tests/integration/safety/test_llama_guard.py

@@ -13,8 +13,8 @@ from collections.abc import Generator
 
 import pytest
 
-from llama_stack.apis.safety import ViolationLevel
 from llama_stack.models.llama.sku_types import CoreModelId
+from llama_stack_api import ViolationLevel
 
 # Llama Guard models available for text and vision shields
 LLAMA_GUARD_TEXT_MODELS = [CoreModelId.llama_guard_4_12b.value]

tests/integration/safety/test_safety.py

@@ -8,7 +8,7 @@ import mimetypes
 
 import pytest
 
-from llama_stack.apis.safety import ViolationLevel
+from llama_stack_api import ViolationLevel
 
 CODE_SCANNER_ENABLED_PROVIDERS = {"ollama", "together", "fireworks"}
 

tests/integration/safety/test_vision_safety.py

@@ -10,7 +10,7 @@ import os
 
 import pytest
 
-from llama_stack.apis.safety import ViolationLevel
+from llama_stack_api import ViolationLevel
 
 VISION_SHIELD_ENABLED_PROVIDERS = {"together"}
 

tests/integration/suites.py

@@ -71,6 +71,26 @@ SETUP_DEFINITIONS: dict[str, Setup] = {
             "embedding_model": "ollama/nomic-embed-text:v1.5",
         },
     ),
+    "ollama-postgres": Setup(
+        name="ollama-postgres",
+        description="Server-mode tests with Postgres-backed persistence",
+        env={
+            "OLLAMA_URL": "http://0.0.0.0:11434",
+            "SAFETY_MODEL": "ollama/llama-guard3:1b",
+            "POSTGRES_HOST": "127.0.0.1",
+            "POSTGRES_PORT": "5432",
+            "POSTGRES_DB": "llamastack",
+            "POSTGRES_USER": "llamastack",
+            "POSTGRES_PASSWORD": "llamastack",
+            "LLAMA_STACK_LOGGING": "openai_responses=info",
+        },
+        defaults={
+            "text_model": "ollama/llama3.2:3b-instruct-fp16",
+            "embedding_model": "sentence-transformers/nomic-embed-text-v1.5",
+            "safety_model": "ollama/llama-guard3:1b",
+            "safety_shield": "llama-guard",
+        },
+    ),
     "vllm": Setup(
         name="vllm",
         description="vLLM provider with a text model",

tests/integration/tool_runtime/test_mcp.py

@@ -4,14 +4,10 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-import json
-
 import pytest
 from llama_stack_client.lib.agents.agent import Agent
 from llama_stack_client.lib.agents.turn_events import StepCompleted, StepProgress, ToolCallIssuedDelta
 
-from llama_stack.core.library_client import LlamaStackAsLibraryClient
-
 AUTH_TOKEN = "test-token"
 
 from tests.common.mcp import MCP_TOOLGROUP_ID, make_mcp_server
@@ -24,9 +20,6 @@ def mcp_server():
 
 
 def test_mcp_invocation(llama_stack_client, text_model_id, mcp_server):
-    if not isinstance(llama_stack_client, LlamaStackAsLibraryClient):
-        pytest.skip("The local MCP server only reliably reachable from library client.")
-
     test_toolgroup_id = MCP_TOOLGROUP_ID
     uri = mcp_server["server_url"]
 
@@ -42,31 +35,20 @@ def test_mcp_invocation(llama_stack_client, text_model_id, mcp_server):
         mcp_endpoint=dict(uri=uri),
     )
 
-    provider_data = {
-        "mcp_headers": {
-            uri: {
-                "Authorization": f"Bearer {AUTH_TOKEN}",
-            },
-        },
-    }
-    auth_headers = {
-        "X-LlamaStack-Provider-Data": json.dumps(provider_data),
-    }
-
-    with pytest.raises(Exception, match="Unauthorized"):
-        llama_stack_client.tools.list(toolgroup_id=test_toolgroup_id)
-
-    tools_list = llama_stack_client.tools.list(
-        toolgroup_id=test_toolgroup_id,
-        extra_headers=auth_headers,
+    # Use the dedicated authorization parameter (no more provider_data headers)
+    # This tests direct tool_runtime.invoke_tool API calls
+    tools_list = llama_stack_client.tool_runtime.list_tools(
+        tool_group_id=test_toolgroup_id,
+        authorization=AUTH_TOKEN,  # Use dedicated authorization parameter
     )
     assert len(tools_list) == 2
     assert {t.name for t in tools_list} == {"greet_everyone", "get_boiling_point"}
 
+    # Invoke tool with authorization parameter
     response = llama_stack_client.tool_runtime.invoke_tool(
         tool_name="greet_everyone",
         kwargs=dict(url="https://www.google.com"),
-        extra_headers=auth_headers,
+        authorization=AUTH_TOKEN,  # Use dedicated authorization parameter
     )
     content = response.content
     assert len(content) == 1
@@ -81,9 +63,7 @@ def test_mcp_invocation(llama_stack_client, text_model_id, mcp_server):
             "server_label": test_toolgroup_id,
             "require_approval": "never",
             "allowed_tools": [tool.name for tool in tools_list],
-            "headers": {
-                "Authorization": f"Bearer {AUTH_TOKEN}",
-            },
+            "authorization": AUTH_TOKEN,
         }
     ]
     agent = Agent(
@@ -109,7 +89,6 @@ def test_mcp_invocation(llama_stack_client, text_model_id, mcp_server):
                 }
             ],
             stream=True,
-            extra_headers=auth_headers,
         )
     )
     events = [chunk.event for chunk in chunks]

tests/integration/tool_runtime/test_mcp_json_schema.py

@@ -4,13 +4,10 @@
 # This source code is licensed under the terms described in the LICENSE file in
 # the root directory of this source tree.
 
-"""
-Integration tests for MCP tools with complex JSON Schema support.
+"""Integration tests for MCP tools with complex JSON Schema support.
 Tests $ref, $defs, and other JSON Schema features through MCP integration.
 """
 
-import json
-
 import pytest
 
 from llama_stack.core.library_client import LlamaStackAsLibraryClient
@@ -123,15 +120,11 @@ class TestMCPSchemaPreservation:
             mcp_endpoint=dict(uri=uri),
         )
 
-        provider_data = {"mcp_headers": {uri: {"Authorization": f"Bearer {AUTH_TOKEN}"}}}
-        auth_headers = {
-            "X-LlamaStack-Provider-Data": json.dumps(provider_data),
-        }
-
+        # Use the dedicated authorization parameter
         # List runtime tools
         response = llama_stack_client.tool_runtime.list_tools(
             tool_group_id=test_toolgroup_id,
-            extra_headers=auth_headers,
+            authorization=AUTH_TOKEN,
         )
 
         tools = response
@@ -166,15 +159,12 @@ class TestMCPSchemaPreservation:
             provider_id="model-context-protocol",
             mcp_endpoint=dict(uri=uri),
         )
-        provider_data = {"mcp_headers": {uri: {"Authorization": f"Bearer {AUTH_TOKEN}"}}}
-        auth_headers = {
-            "X-LlamaStack-Provider-Data": json.dumps(provider_data),
-        }
 
+        # Use the dedicated authorization parameter
         # List tools
         response = llama_stack_client.tool_runtime.list_tools(
             tool_group_id=test_toolgroup_id,
-            extra_headers=auth_headers,
+            authorization=AUTH_TOKEN,
         )
 
         # Find book_flight tool (which should have $ref/$defs)
@@ -216,14 +206,10 @@ class TestMCPSchemaPreservation:
             mcp_endpoint=dict(uri=uri),
         )
 
-        provider_data = {"mcp_headers": {uri: {"Authorization": f"Bearer {AUTH_TOKEN}"}}}
-        auth_headers = {
-            "X-LlamaStack-Provider-Data": json.dumps(provider_data),
-        }
-
+        # Use the dedicated authorization parameter
         response = llama_stack_client.tool_runtime.list_tools(
             tool_group_id=test_toolgroup_id,
-            extra_headers=auth_headers,
+            authorization=AUTH_TOKEN,
         )
 
         # Find get_weather tool
@@ -263,15 +249,10 @@ class TestMCPToolInvocation:
             mcp_endpoint=dict(uri=uri),
         )
 
-        provider_data = {"mcp_headers": {uri: {"Authorization": f"Bearer {AUTH_TOKEN}"}}}
-        auth_headers = {
-            "X-LlamaStack-Provider-Data": json.dumps(provider_data),
-        }
-
-        # List tools to populate the tool index
+        # Use the dedicated authorization parameter
         llama_stack_client.tool_runtime.list_tools(
             tool_group_id=test_toolgroup_id,
-            extra_headers=auth_headers,
+            authorization=AUTH_TOKEN,
         )
 
         # Invoke tool with complex nested data
@@ -283,7 +264,7 @@ class TestMCPToolInvocation:
                     "shipping": {"address": {"street": "123 Main St", "city": "San Francisco", "zipcode": "94102"}},
                 }
             },
-            extra_headers=auth_headers,
+            authorization=AUTH_TOKEN,
         )
 
         # Should succeed without schema validation errors
@@ -309,22 +290,17 @@ class TestMCPToolInvocation:
             mcp_endpoint=dict(uri=uri),
         )
 
-        provider_data = {"mcp_headers": {uri: {"Authorization": f"Bearer {AUTH_TOKEN}"}}}
-        auth_headers = {
-            "X-LlamaStack-Provider-Data": json.dumps(provider_data),
-        }
-
-        # List tools to populate the tool index
+        # Use the dedicated authorization parameter
         llama_stack_client.tool_runtime.list_tools(
             tool_group_id=test_toolgroup_id,
-            extra_headers=auth_headers,
+            authorization=AUTH_TOKEN,
         )
 
         # Test with email format
         result_email = llama_stack_client.tool_runtime.invoke_tool(
             tool_name="flexible_contact",
             kwargs={"contact_info": "user@example.com"},
-            extra_headers=auth_headers,
+            authorization=AUTH_TOKEN,
         )
 
         assert result_email.error_message is None
@@ -333,7 +309,7 @@ class TestMCPToolInvocation:
         result_phone = llama_stack_client.tool_runtime.invoke_tool(
             tool_name="flexible_contact",
             kwargs={"contact_info": "+15551234567"},
-            extra_headers=auth_headers,
+            authorization=AUTH_TOKEN,
         )
 
         assert result_phone.error_message is None
@@ -365,14 +341,10 @@ class TestAgentWithMCPTools:
             mcp_endpoint=dict(uri=uri),
         )
 
-        provider_data = {"mcp_headers": {uri: {"Authorization": f"Bearer {AUTH_TOKEN}"}}}
-        auth_headers = {
-            "X-LlamaStack-Provider-Data": json.dumps(provider_data),
-        }
-
-        tools_list = llama_stack_client.tools.list(
-            toolgroup_id=test_toolgroup_id,
-            extra_headers=auth_headers,
+        # Use the dedicated authorization parameter
+        tools_list = llama_stack_client.tool_runtime.list_tools(
+            tool_group_id=test_toolgroup_id,
+            authorization=AUTH_TOKEN,
         )
         tool_defs = [
             {
@@ -381,6 +353,7 @@ class TestAgentWithMCPTools:
                 "server_label": test_toolgroup_id,
                 "require_approval": "never",
                 "allowed_tools": [tool.name for tool in tools_list],
+                "authorization": AUTH_TOKEN,
             }
         ]
 
@@ -389,7 +362,6 @@ class TestAgentWithMCPTools:
             model=text_model_id,
             instructions="You are a helpful assistant that can process orders and book flights.",
             tools=tool_defs,
-            extra_headers=auth_headers,
         )
 
         session_id = agent.create_session("test-session-complex")
@@ -411,7 +383,6 @@ class TestAgentWithMCPTools:
                     }
                 ],
                 stream=True,
-                extra_headers=auth_headers,
             )
         )
 

tests/integration/tool_runtime/test_registration.py

@@ -8,8 +8,8 @@ import re
 
 import pytest
 
-from llama_stack.apis.common.errors import ToolGroupNotFoundError
 from llama_stack.core.library_client import LlamaStackAsLibraryClient
+from llama_stack_api import ToolGroupNotFoundError
 from tests.common.mcp import MCP_TOOLGROUP_ID, make_mcp_server
 
 

tests/integration/vector_io/test_openai_vector_stores.py

@@ -11,9 +11,9 @@ import pytest
 from llama_stack_client import BadRequestError
 from openai import BadRequestError as OpenAIBadRequestError
 
-from llama_stack.apis.vector_io import Chunk
 from llama_stack.core.library_client import LlamaStackAsLibraryClient
 from llama_stack.log import get_logger
+from llama_stack_api import Chunk, ExpiresAfter
 
 from ..conftest import vector_provider_wrapper
 
@@ -649,7 +649,7 @@ def test_openai_vector_store_attach_file(
 ):
     """Test OpenAI vector store attach file."""
     skip_if_provider_doesnt_support_openai_vector_stores(client_with_models)
-    from llama_stack.apis.files import ExpiresAfter
+    from llama_stack_api import ExpiresAfter
 
     compat_client = compat_client_with_empty_stores
 
@@ -713,7 +713,7 @@ def test_openai_vector_store_attach_files_on_creation(
     skip_if_provider_doesnt_support_openai_vector_stores(client_with_models)
 
     compat_client = compat_client_with_empty_stores
-    from llama_stack.apis.files import ExpiresAfter
+    from llama_stack_api import ExpiresAfter
 
     # Create some files and attach them to the vector store
     valid_file_ids = []
@@ -778,7 +778,7 @@ def test_openai_vector_store_list_files(
     skip_if_provider_doesnt_support_openai_vector_stores(client_with_models)
 
     compat_client = compat_client_with_empty_stores
-    from llama_stack.apis.files import ExpiresAfter
+    from llama_stack_api import ExpiresAfter
 
     # Create a vector store
     vector_store = compat_client.vector_stores.create(
@@ -870,7 +870,7 @@ def test_openai_vector_store_retrieve_file_contents(
     skip_if_provider_doesnt_support_openai_vector_stores(client_with_models)
 
     compat_client = compat_client_with_empty_stores
-    from llama_stack.apis.files import ExpiresAfter
+    from llama_stack_api import ExpiresAfter
 
     # Create a vector store
     vector_store = compat_client.vector_stores.create(
@@ -911,16 +911,16 @@ def test_openai_vector_store_retrieve_file_contents(
     )
 
     assert file_contents is not None
-    assert len(file_contents.content) == 1
-    content = file_contents.content[0]
+    assert file_contents.object == "vector_store.file_content.page"
+    assert len(file_contents.data) == 1
+    content = file_contents.data[0]
 
     # llama-stack-client returns a model, openai-python is a badboy and returns a dict
     if not isinstance(content, dict):
         content = content.model_dump()
     assert content["type"] == "text"
     assert content["text"] == test_content.decode("utf-8")
-    assert file_contents.filename == file_name
-    assert file_contents.attributes == attributes
+    assert file_contents.has_more is False
 
 
 @vector_provider_wrapper
@@ -931,7 +931,7 @@ def test_openai_vector_store_delete_file(
     skip_if_provider_doesnt_support_openai_vector_stores(client_with_models)
 
     compat_client = compat_client_with_empty_stores
-    from llama_stack.apis.files import ExpiresAfter
+    from llama_stack_api import ExpiresAfter
 
     # Create a vector store
     vector_store = compat_client.vector_stores.create(
@@ -997,7 +997,7 @@ def test_openai_vector_store_delete_file_removes_from_vector_store(
     skip_if_provider_doesnt_support_openai_vector_stores(client_with_models)
 
     compat_client = compat_client_with_empty_stores
-    from llama_stack.apis.files import ExpiresAfter
+    from llama_stack_api import ExpiresAfter
 
     # Create a vector store
     vector_store = compat_client.vector_stores.create(
@@ -1049,7 +1049,7 @@ def test_openai_vector_store_update_file(
     skip_if_provider_doesnt_support_openai_vector_stores(client_with_models)
 
     compat_client = compat_client_with_empty_stores
-    from llama_stack.apis.files import ExpiresAfter
+    from llama_stack_api import ExpiresAfter
 
     # Create a vector store
     vector_store = compat_client.vector_stores.create(
@@ -1106,7 +1106,7 @@ def test_create_vector_store_files_duplicate_vector_store_name(
     This test confirms that client.vector_stores.create() creates a unique ID
     """
     skip_if_provider_doesnt_support_openai_vector_stores(client_with_models)
-    from llama_stack.apis.files import ExpiresAfter
+    from llama_stack_api import ExpiresAfter
 
     compat_client = compat_client_with_empty_stores
 
@@ -1487,14 +1487,12 @@ def test_openai_vector_store_file_batch_retrieve_contents(
         )
 
         assert file_contents is not None
-        assert file_contents.filename == file_data[i][0]
-        assert len(file_contents.content) > 0
+        assert file_contents.object == "vector_store.file_content.page"
+        assert len(file_contents.data) > 0
 
         # Verify the content matches what we uploaded
         content_text = (
-            file_contents.content[0].text
-            if hasattr(file_contents.content[0], "text")
-            else file_contents.content[0]["text"]
+            file_contents.data[0].text if hasattr(file_contents.data[0], "text") else file_contents.data[0]["text"]
         )
         assert file_data[i][1].decode("utf-8") in content_text
 
@@ -1610,3 +1608,97 @@ def test_openai_vector_store_embedding_config_from_metadata(
 
     assert "metadata_config_store" in store_names
     assert "consistent_config_store" in store_names
+
+
+@vector_provider_wrapper
+def test_openai_vector_store_file_contents_with_extra_query(
+    compat_client_with_empty_stores, client_with_models, embedding_model_id, embedding_dimension, vector_io_provider_id
+):
+    """Test that vector store file contents endpoint supports extra_query parameter."""
+    skip_if_provider_doesnt_support_openai_vector_stores(client_with_models)
+    compat_client = compat_client_with_empty_stores
+
+    # Create a vector store
+    vector_store = compat_client.vector_stores.create(
+        name="test_extra_query_store",
+        extra_body={
+            "embedding_model": embedding_model_id,
+            "provider_id": vector_io_provider_id,
+        },
+    )
+
+    # Create and attach a file
+    test_content = b"This is test content for extra_query validation."
+    with BytesIO(test_content) as file_buffer:
+        file_buffer.name = "test_extra_query.txt"
+        file = compat_client.files.create(
+            file=file_buffer,
+            purpose="assistants",
+            expires_after=ExpiresAfter(anchor="created_at", seconds=86400),
+        )
+
+    file_attach_response = compat_client.vector_stores.files.create(
+        vector_store_id=vector_store.id,
+        file_id=file.id,
+        extra_body={"embedding_model": embedding_model_id},
+    )
+    assert file_attach_response.status == "completed"
+
+    # Wait for processing
+    time.sleep(2)
+
+    # Test that extra_query parameter is accepted and processed
+    content_with_extra_query = compat_client.vector_stores.files.content(
+        vector_store_id=vector_store.id,
+        file_id=file.id,
+        extra_query={"include_embeddings": True, "include_metadata": True},
+    )
+
+    # Test without extra_query for comparison
+    content_without_extra_query = compat_client.vector_stores.files.content(
+        vector_store_id=vector_store.id,
+        file_id=file.id,
+    )
+
+    # Validate that both calls succeed
+    assert content_with_extra_query is not None
+    assert content_without_extra_query is not None
+    assert len(content_with_extra_query.data) > 0
+    assert len(content_without_extra_query.data) > 0
+
+    # Validate that extra_query parameter is processed correctly
+    # Both should have the embedding/metadata fields available (may be None based on flags)
+    first_chunk_with_flags = content_with_extra_query.data[0]
+    first_chunk_without_flags = content_without_extra_query.data[0]
+
+    # The key validation: extra_query fields are present in the response
+    # Handle both dict and object responses (different clients may return different formats)
+    def has_field(obj, field):
+        if isinstance(obj, dict):
+            return field in obj
+        else:
+            return hasattr(obj, field)
+
+    # Validate that all expected fields are present in both responses
+    expected_fields = ["embedding", "chunk_metadata", "metadata", "text"]
+    for field in expected_fields:
+        assert has_field(first_chunk_with_flags, field), f"Field '{field}' missing from response with extra_query"
+        assert has_field(first_chunk_without_flags, field), f"Field '{field}' missing from response without extra_query"
+
+    # Validate content is the same
+    def get_field(obj, field):
+        if isinstance(obj, dict):
+            return obj[field]
+        else:
+            return getattr(obj, field)
+
+    assert get_field(first_chunk_with_flags, "text") == test_content.decode("utf-8")
+    assert get_field(first_chunk_without_flags, "text") == test_content.decode("utf-8")
+
+    with_flags_embedding = get_field(first_chunk_with_flags, "embedding")
+    without_flags_embedding = get_field(first_chunk_without_flags, "embedding")
+
+    # Validate that embeddings are included when requested and excluded when not requested
+    assert with_flags_embedding is not None, "Embeddings should be included when include_embeddings=True"
+    assert len(with_flags_embedding) > 0, "Embedding should be a non-empty list"
+    assert without_flags_embedding is None, "Embeddings should not be included when include_embeddings=False"

tests/integration/vector_io/test_vector_io.py

@@ -6,7 +6,7 @@
 
 import pytest
 
-from llama_stack.apis.vector_io import Chunk
+from llama_stack_api import Chunk
 
 from ..conftest import vector_provider_wrapper
 

tests/unit/conversations/test_api_models.py

@@ -5,11 +5,7 @@
 # the root directory of this source tree.
 
 
-from llama_stack.apis.conversations.conversations import (
-    Conversation,
-    ConversationItem,
-    ConversationItemList,
-)
+from llama_stack_api import Conversation, ConversationItem, ConversationItemList
 
 
 def test_conversation_model_defaults():

tests/unit/conversations/test_conversations.py

@@ -12,10 +12,6 @@ from openai.types.conversations.conversation import Conversation as OpenAIConver
 from openai.types.conversations.conversation_item import ConversationItem as OpenAIConversationItem
 from pydantic import TypeAdapter
 
-from llama_stack.apis.agents.openai_responses import (
-    OpenAIResponseInputMessageContentText,
-    OpenAIResponseMessage,
-)
 from llama_stack.core.conversations.conversations import (
     ConversationServiceConfig,
     ConversationServiceImpl,
@@ -27,7 +23,8 @@ from llama_stack.core.storage.datatypes import (
     SqlStoreReference,
     StorageConfig,
 )
-from llama_stack.providers.utils.sqlstore.sqlstore import register_sqlstore_backends
+from llama_stack.core.storage.sqlstore.sqlstore import register_sqlstore_backends
+from llama_stack_api import OpenAIResponseInputMessageContentText, OpenAIResponseMessage
 
 
 @pytest.fixture
@@ -41,6 +38,9 @@ async def service():
             },
             stores=ServerStoresConfig(
                 conversations=SqlStoreReference(backend="sql_test", table_name="openai_conversations"),
+                metadata=None,
+                inference=None,
+                prompts=None,
             ),
         )
         register_sqlstore_backends({"sql_test": storage.backends["sql_test"]})
@@ -145,6 +145,9 @@ async def test_policy_configuration():
             },
             stores=ServerStoresConfig(
                 conversations=SqlStoreReference(backend="sql_test", table_name="openai_conversations"),
+                metadata=None,
+                inference=None,
+                prompts=None,
             ),
         )
         register_sqlstore_backends({"sql_test": storage.backends["sql_test"]})

tests/unit/core/routers/test_safety_router.py

@@ -6,10 +6,9 @@
 
 from unittest.mock import AsyncMock
 
-from llama_stack.apis.safety.safety import ModerationObject, ModerationObjectResults
-from llama_stack.apis.shields import ListShieldsResponse, Shield
 from llama_stack.core.datatypes import SafetyConfig
 from llama_stack.core.routers.safety import SafetyRouter
+from llama_stack_api import ListShieldsResponse, ModerationObject, ModerationObjectResults, Shield
 
 
 async def test_run_moderation_uses_default_shield_when_model_missing():

tests/unit/core/routers/test_vector_io.py

@@ -8,8 +8,13 @@ from unittest.mock import AsyncMock, Mock
 
 import pytest
 
-from llama_stack.apis.vector_io import OpenAICreateVectorStoreRequestWithExtraBody
 from llama_stack.core.routers.vector_io import VectorIORouter
+from llama_stack_api import (
+    ModelNotFoundError,
+    ModelType,
+    ModelTypeError,
+    OpenAICreateVectorStoreRequestWithExtraBody,
+)
 
 
 async def test_single_provider_auto_selection():
@@ -21,6 +26,7 @@ async def test_single_provider_auto_selection():
             Mock(identifier="all-MiniLM-L6-v2", model_type="embedding", metadata={"embedding_dimension": 384})
         ]
     )
+    mock_routing_table.get_object_by_identifier = AsyncMock(return_value=Mock(model_type=ModelType.embedding))
     mock_routing_table.register_vector_store = AsyncMock(
         return_value=Mock(identifier="vs_123", provider_id="inline::faiss", provider_resource_id="vs_123")
     )
@@ -48,6 +54,7 @@ async def test_create_vector_stores_multiple_providers_missing_provider_id_error
             Mock(identifier="all-MiniLM-L6-v2", model_type="embedding", metadata={"embedding_dimension": 384})
         ]
     )
+    mock_routing_table.get_object_by_identifier = AsyncMock(return_value=Mock(model_type=ModelType.embedding))
     router = VectorIORouter(mock_routing_table)
     request = OpenAICreateVectorStoreRequestWithExtraBody.model_validate(
         {"name": "test_store", "embedding_model": "all-MiniLM-L6-v2"}
@@ -55,3 +62,94 @@ async def test_create_vector_stores_multiple_providers_missing_provider_id_error
 
     with pytest.raises(ValueError, match="Multiple vector_io providers available"):
         await router.openai_create_vector_store(request)
+
+
+async def test_update_vector_store_provider_id_change_fails():
+    """Test that updating a vector store with a different provider_id fails with clear error."""
+    mock_routing_table = Mock()
+
+    # Mock an existing vector store with provider_id "faiss"
+    mock_existing_store = Mock()
+    mock_existing_store.provider_id = "inline::faiss"
+    mock_existing_store.identifier = "vs_123"
+
+    mock_routing_table.get_object_by_identifier = AsyncMock(return_value=mock_existing_store)
+    mock_routing_table.get_provider_impl = AsyncMock(
+        return_value=Mock(openai_update_vector_store=AsyncMock(return_value=Mock(id="vs_123")))
+    )
+
+    router = VectorIORouter(mock_routing_table)
+
+    # Try to update with different provider_id in metadata - this should fail
+    with pytest.raises(ValueError, match="provider_id cannot be changed after vector store creation"):
+        await router.openai_update_vector_store(
+            vector_store_id="vs_123",
+            name="updated_name",
+            metadata={"provider_id": "inline::sqlite"},  # Different provider_id
+        )
+
+    # Verify the existing store was looked up to check provider_id
+    mock_routing_table.get_object_by_identifier.assert_called_once_with("vector_store", "vs_123")
+
+    # Provider should not be called since validation failed
+    mock_routing_table.get_provider_impl.assert_not_called()
+
+
+async def test_update_vector_store_same_provider_id_succeeds():
+    """Test that updating a vector store with the same provider_id succeeds."""
+    mock_routing_table = Mock()
+
+    # Mock an existing vector store with provider_id "faiss"
+    mock_existing_store = Mock()
+    mock_existing_store.provider_id = "inline::faiss"
+    mock_existing_store.identifier = "vs_123"
+
+    mock_routing_table.get_object_by_identifier = AsyncMock(return_value=mock_existing_store)
+    mock_routing_table.get_provider_impl = AsyncMock(
+        return_value=Mock(openai_update_vector_store=AsyncMock(return_value=Mock(id="vs_123")))
+    )
+
+    router = VectorIORouter(mock_routing_table)
+
+    # Update with same provider_id should succeed
+    await router.openai_update_vector_store(
+        vector_store_id="vs_123",
+        name="updated_name",
+        metadata={"provider_id": "inline::faiss"},  # Same provider_id
+    )
+
+    # Verify the provider update method was called
+    mock_routing_table.get_provider_impl.assert_called_once_with("vs_123")
+    provider = await mock_routing_table.get_provider_impl("vs_123")
+    provider.openai_update_vector_store.assert_called_once_with(
+        vector_store_id="vs_123", name="updated_name", expires_after=None, metadata={"provider_id": "inline::faiss"}
+    )
+
+
+async def test_create_vector_store_with_unknown_embedding_model_raises_error():
+    """Test that creating a vector store with an unknown embedding model raises
+    FoundError."""
+    mock_routing_table = Mock(impls_by_provider_id={"provider": "mock"})
+    mock_routing_table.get_object_by_identifier = AsyncMock(return_value=None)
+
+    router = VectorIORouter(mock_routing_table)
+    request = OpenAICreateVectorStoreRequestWithExtraBody.model_validate(
+        {"embedding_model": "unknown-model", "embedding_dimension": 384}
+    )
+
+    with pytest.raises(ModelNotFoundError, match="Model 'unknown-model' not found"):
+        await router.openai_create_vector_store(request)
+
+
+async def test_create_vector_store_with_wrong_model_type_raises_error():
+    """Test that creating a vector store with a non-embedding model raises ModelTypeError."""
+    mock_routing_table = Mock(impls_by_provider_id={"provider": "mock"})
+    mock_routing_table.get_object_by_identifier = AsyncMock(return_value=Mock(model_type=ModelType.llm))
+
+    router = VectorIORouter(mock_routing_table)
+    request = OpenAICreateVectorStoreRequestWithExtraBody.model_validate(
+        {"embedding_model": "text-model", "embedding_dimension": 384}
+    )
+
+    with pytest.raises(ModelTypeError, match="Model 'text-model' is of type"):
+        await router.openai_create_vector_store(request)

tests/unit/core/test_stack_validation.py

@@ -10,11 +10,10 @@ from unittest.mock import AsyncMock
 
 import pytest
 
-from llama_stack.apis.models import ListModelsResponse, Model, ModelType
-from llama_stack.apis.shields import ListShieldsResponse, Shield
-from llama_stack.core.datatypes import QualifiedModel, SafetyConfig, StackRunConfig, StorageConfig, VectorStoresConfig
+from llama_stack.core.datatypes import QualifiedModel, SafetyConfig, StackRunConfig, VectorStoresConfig
 from llama_stack.core.stack import validate_safety_config, validate_vector_stores_config
-from llama_stack.providers.datatypes import Api
+from llama_stack.core.storage.datatypes import ServerStoresConfig, StorageConfig
+from llama_stack_api import Api, ListModelsResponse, ListShieldsResponse, Model, ModelType, Shield
 
 
 class TestVectorStoresValidation:
@@ -23,7 +22,15 @@ class TestVectorStoresValidation:
         run_config = StackRunConfig(
             image_name="test",
             providers={},
-            storage=StorageConfig(backends={}, stores={}),
+            storage=StorageConfig(
+                backends={},
+                stores=ServerStoresConfig(
+                    metadata=None,
+                    inference=None,
+                    conversations=None,
+                    prompts=None,
+                ),
+            ),
             vector_stores=VectorStoresConfig(
                 default_provider_id="faiss",
                 default_embedding_model=QualifiedModel(
@@ -43,7 +50,15 @@ class TestVectorStoresValidation:
         run_config = StackRunConfig(
             image_name="test",
             providers={},
-            storage=StorageConfig(backends={}, stores={}),
+            storage=StorageConfig(
+                backends={},
+                stores=ServerStoresConfig(
+                    metadata=None,
+                    inference=None,
+                    conversations=None,
+                    prompts=None,
+                ),
+            ),
             vector_stores=VectorStoresConfig(
                 default_provider_id="faiss",
                 default_embedding_model=QualifiedModel(

tests/unit/distribution/routers/test_routing_tables.py

@@ -10,14 +10,6 @@ from unittest.mock import AsyncMock
 
 import pytest
 
-from llama_stack.apis.common.content_types import URL
-from llama_stack.apis.common.errors import ModelNotFoundError
-from llama_stack.apis.common.type_system import NumberType
-from llama_stack.apis.datasets.datasets import Dataset, DatasetPurpose, URIDataSource
-from llama_stack.apis.datatypes import Api
-from llama_stack.apis.models import Model, ModelType
-from llama_stack.apis.shields.shields import Shield
-from llama_stack.apis.tools import ListToolDefsResponse, ToolDef, ToolGroup
 from llama_stack.core.datatypes import RegistryEntrySource
 from llama_stack.core.routing_tables.benchmarks import BenchmarksRoutingTable
 from llama_stack.core.routing_tables.datasets import DatasetsRoutingTable
@@ -25,6 +17,21 @@ from llama_stack.core.routing_tables.models import ModelsRoutingTable
 from llama_stack.core.routing_tables.scoring_functions import ScoringFunctionsRoutingTable
 from llama_stack.core.routing_tables.shields import ShieldsRoutingTable
 from llama_stack.core.routing_tables.toolgroups import ToolGroupsRoutingTable
+from llama_stack_api import (
+    URL,
+    Api,
+    Dataset,
+    DatasetPurpose,
+    ListToolDefsResponse,
+    Model,
+    ModelNotFoundError,
+    ModelType,
+    NumberType,
+    Shield,
+    ToolDef,
+    ToolGroup,
+    URIDataSource,
+)
 
 
 class Impl:
@@ -130,7 +137,7 @@ class ToolGroupsImpl(Impl):
     async def unregister_toolgroup(self, toolgroup_id: str):
         return toolgroup_id
 
-    async def list_runtime_tools(self, toolgroup_id, mcp_endpoint):
+    async def list_runtime_tools(self, toolgroup_id, mcp_endpoint, authorization=None):
         return ListToolDefsResponse(
             data=[
                 ToolDef(

tests/unit/distribution/test_api_recordings.py

@@ -11,8 +11,15 @@ from unittest.mock import patch
 import pytest
 from openai import AsyncOpenAI
 
+from llama_stack.testing.api_recorder import (
+    APIRecordingMode,
+    ResponseStorage,
+    api_recording,
+    normalize_inference_request,
+)
+
 # Import the real Pydantic response types instead of using Mocks
-from llama_stack.apis.inference import (
+from llama_stack_api import (
     OpenAIAssistantMessageParam,
     OpenAIChatCompletion,
     OpenAIChoice,
@@ -20,12 +27,6 @@ from llama_stack.apis.inference import (
     OpenAIEmbeddingsResponse,
     OpenAIEmbeddingUsage,
 )
-from llama_stack.testing.api_recorder import (
-    APIRecordingMode,
-    ResponseStorage,
-    api_recording,
-    normalize_inference_request,
-)
 
 
 @pytest.fixture

tests/unit/distribution/test_distribution.py

@@ -22,7 +22,7 @@ from llama_stack.core.storage.datatypes import (
     SqlStoreReference,
     StorageConfig,
 )
-from llama_stack.providers.datatypes import ProviderSpec
+from llama_stack_api import ProviderSpec
 
 
 class SampleConfig(BaseModel):
@@ -312,7 +312,7 @@ pip_packages:
         """Test loading an external provider from a module (success path)."""
         from types import SimpleNamespace
 
-        from llama_stack.providers.datatypes import Api, ProviderSpec
+        from llama_stack_api import Api, ProviderSpec
 
         # Simulate a provider module with get_provider_spec
         fake_spec = ProviderSpec(
@@ -396,7 +396,7 @@ pip_packages:
     def test_external_provider_from_module_building(self, mock_providers):
         """Test loading an external provider from a module during build (building=True, partial spec)."""
         from llama_stack.core.datatypes import BuildConfig, BuildProvider, DistributionSpec
-        from llama_stack.providers.datatypes import Api
+        from llama_stack_api import Api
 
         # No importlib patch needed, should not import module when type of `config` is BuildConfig or DistributionSpec
         build_config = BuildConfig(
@@ -457,7 +457,7 @@ class TestGetExternalProvidersFromModule:
         from types import SimpleNamespace
 
         from llama_stack.core.distribution import get_external_providers_from_module
-        from llama_stack.providers.datatypes import ProviderSpec
+        from llama_stack_api import ProviderSpec
 
         fake_spec = ProviderSpec(
             api=Api.inference,
@@ -594,7 +594,7 @@ class TestGetExternalProvidersFromModule:
         from types import SimpleNamespace
 
         from llama_stack.core.distribution import get_external_providers_from_module
-        from llama_stack.providers.datatypes import ProviderSpec
+        from llama_stack_api import ProviderSpec
 
         spec1 = ProviderSpec(
             api=Api.inference,
@@ -642,7 +642,7 @@ class TestGetExternalProvidersFromModule:
         from types import SimpleNamespace
 
         from llama_stack.core.distribution import get_external_providers_from_module
-        from llama_stack.providers.datatypes import ProviderSpec
+        from llama_stack_api import ProviderSpec
 
         spec1 = ProviderSpec(
             api=Api.inference,
@@ -690,7 +690,7 @@ class TestGetExternalProvidersFromModule:
         from types import SimpleNamespace
 
         from llama_stack.core.distribution import get_external_providers_from_module
-        from llama_stack.providers.datatypes import ProviderSpec
+        from llama_stack_api import ProviderSpec
 
         # Module returns both inline and remote variants
         spec1 = ProviderSpec(
@@ -829,7 +829,7 @@ class TestGetExternalProvidersFromModule:
         from types import SimpleNamespace
 
         from llama_stack.core.distribution import get_external_providers_from_module
-        from llama_stack.providers.datatypes import ProviderSpec
+        from llama_stack_api import ProviderSpec
 
         inference_spec = ProviderSpec(
             api=Api.inference,

tests/unit/files/test_files.py

@@ -7,16 +7,14 @@
 
 import pytest
 
-from llama_stack.apis.common.errors import ResourceNotFoundError
-from llama_stack.apis.common.responses import Order
-from llama_stack.apis.files import OpenAIFilePurpose
 from llama_stack.core.access_control.access_control import default_policy
 from llama_stack.core.storage.datatypes import SqliteSqlStoreConfig, SqlStoreReference
+from llama_stack.core.storage.sqlstore.sqlstore import register_sqlstore_backends
 from llama_stack.providers.inline.files.localfs import (
     LocalfsFilesImpl,
     LocalfsFilesImplConfig,
 )
-from llama_stack.providers.utils.sqlstore.sqlstore import register_sqlstore_backends
+from llama_stack_api import OpenAIFilePurpose, Order, ResourceNotFoundError
 
 
 class MockUploadFile:

tests/unit/fixtures.py

@@ -6,9 +6,9 @@
 
 import pytest
 
+from llama_stack.core.storage.kvstore.config import SqliteKVStoreConfig
+from llama_stack.core.storage.kvstore.sqlite import SqliteKVStoreImpl
 from llama_stack.core.store.registry import CachedDiskDistributionRegistry, DiskDistributionRegistry
-from llama_stack.providers.utils.kvstore.config import SqliteKVStoreConfig
-from llama_stack.providers.utils.kvstore.sqlite import SqliteKVStoreImpl
 
 
 @pytest.fixture(scope="function")

tests/unit/models/test_prompt_adapter.py

@@ -1,303 +0,0 @@
-# Copyright (c) Meta Platforms, Inc. and affiliates.
-# All rights reserved.
-#
-# This source code is licensed under the terms described in the LICENSE file in
-# the root directory of this source tree.
-
-
-from llama_stack.apis.inference import (
-    ChatCompletionRequest,
-    CompletionMessage,
-    StopReason,
-    SystemMessage,
-    SystemMessageBehavior,
-    ToolCall,
-    ToolConfig,
-    UserMessage,
-)
-from llama_stack.models.llama.datatypes import (
-    BuiltinTool,
-    ToolDefinition,
-    ToolPromptFormat,
-)
-from llama_stack.providers.utils.inference.prompt_adapter import (
-    chat_completion_request_to_messages,
-    chat_completion_request_to_prompt,
-    interleaved_content_as_str,
-)
-
-MODEL = "Llama3.1-8B-Instruct"
-MODEL3_2 = "Llama3.2-3B-Instruct"
-
-
-async def test_system_default():
-    content = "Hello !"
-    request = ChatCompletionRequest(
-        model=MODEL,
-        messages=[
-            UserMessage(content=content),
-        ],
-    )
-    messages = chat_completion_request_to_messages(request, MODEL)
-    assert len(messages) == 2
-    assert messages[-1].content == content
-    assert "Cutting Knowledge Date: December 2023" in interleaved_content_as_str(messages[0].content)
-
-
-async def test_system_builtin_only():
-    content = "Hello !"
-    request = ChatCompletionRequest(
-        model=MODEL,
-        messages=[
-            UserMessage(content=content),
-        ],
-        tools=[
-            ToolDefinition(tool_name=BuiltinTool.code_interpreter),
-            ToolDefinition(tool_name=BuiltinTool.brave_search),
-        ],
-    )
-    messages = chat_completion_request_to_messages(request, MODEL)
-    assert len(messages) == 2
-    assert messages[-1].content == content
-    assert "Cutting Knowledge Date: December 2023" in interleaved_content_as_str(messages[0].content)
-    assert "Tools: brave_search" in interleaved_content_as_str(messages[0].content)
-
-
-async def test_system_custom_only():
-    content = "Hello !"
-    request = ChatCompletionRequest(
-        model=MODEL,
-        messages=[
-            UserMessage(content=content),
-        ],
-        tools=[
-            ToolDefinition(
-                tool_name="custom1",
-                description="custom1 tool",
-                input_schema={
-                    "type": "object",
-                    "properties": {
-                        "param1": {
-                            "type": "str",
-                            "description": "param1 description",
-                        },
-                    },
-                    "required": ["param1"],
-                },
-            )
-        ],
-        tool_config=ToolConfig(tool_prompt_format=ToolPromptFormat.json),
-    )
-    messages = chat_completion_request_to_messages(request, MODEL)
-    assert len(messages) == 3
-    assert "Environment: ipython" in interleaved_content_as_str(messages[0].content)
-
-    assert "Return function calls in JSON format" in interleaved_content_as_str(messages[1].content)
-    assert messages[-1].content == content
-
-
-async def test_system_custom_and_builtin():
-    content = "Hello !"
-    request = ChatCompletionRequest(
-        model=MODEL,
-        messages=[
-            UserMessage(content=content),
-        ],
-        tools=[
-            ToolDefinition(tool_name=BuiltinTool.code_interpreter),
-            ToolDefinition(tool_name=BuiltinTool.brave_search),
-            ToolDefinition(
-                tool_name="custom1",
-                description="custom1 tool",
-                input_schema={
-                    "type": "object",
-                    "properties": {
-                        "param1": {
-                            "type": "str",
-                            "description": "param1 description",
-                        },
-                    },
-                    "required": ["param1"],
-                },
-            ),
-        ],
-    )
-    messages = chat_completion_request_to_messages(request, MODEL)
-    assert len(messages) == 3
-
-    assert "Environment: ipython" in interleaved_content_as_str(messages[0].content)
-    assert "Tools: brave_search" in interleaved_content_as_str(messages[0].content)
-
-    assert "Return function calls in JSON format" in interleaved_content_as_str(messages[1].content)
-    assert messages[-1].content == content
-
-
-async def test_completion_message_encoding():
-    request = ChatCompletionRequest(
-        model=MODEL3_2,
-        messages=[
-            UserMessage(content="hello"),
-            CompletionMessage(
-                content="",
-                stop_reason=StopReason.end_of_turn,
-                tool_calls=[
-                    ToolCall(
-                        tool_name="custom1",
-                        arguments='{"param1": "value1"}',  # arguments must be a JSON string
-                        call_id="123",
-                    )
-                ],
-            ),
-        ],
-        tools=[
-            ToolDefinition(
-                tool_name="custom1",
-                description="custom1 tool",
-                input_schema={
-                    "type": "object",
-                    "properties": {
-                        "param1": {
-                            "type": "str",
-                            "description": "param1 description",
-                        },
-                    },
-                    "required": ["param1"],
-                },
-            ),
-        ],
-        tool_config=ToolConfig(tool_prompt_format=ToolPromptFormat.python_list),
-    )
-    prompt = await chat_completion_request_to_prompt(request, request.model)
-    assert '[custom1(param1="value1")]' in prompt
-
-    request.model = MODEL
-    request.tool_config = ToolConfig(tool_prompt_format=ToolPromptFormat.json)
-    prompt = await chat_completion_request_to_prompt(request, request.model)
-    assert '{"type": "function", "name": "custom1", "parameters": {"param1": "value1"}}' in prompt
-
-
-async def test_user_provided_system_message():
-    content = "Hello !"
-    system_prompt = "You are a pirate"
-    request = ChatCompletionRequest(
-        model=MODEL,
-        messages=[
-            SystemMessage(content=system_prompt),
-            UserMessage(content=content),
-        ],
-        tools=[
-            ToolDefinition(tool_name=BuiltinTool.code_interpreter),
-        ],
-    )
-    messages = chat_completion_request_to_messages(request, MODEL)
-    assert len(messages) == 2
-    assert interleaved_content_as_str(messages[0].content).endswith(system_prompt)
-
-    assert messages[-1].content == content
-
-
-async def test_replace_system_message_behavior_builtin_tools():
-    content = "Hello !"
-    system_prompt = "You are a pirate"
-    request = ChatCompletionRequest(
-        model=MODEL,
-        messages=[
-            SystemMessage(content=system_prompt),
-            UserMessage(content=content),
-        ],
-        tools=[
-            ToolDefinition(tool_name=BuiltinTool.code_interpreter),
-        ],
-        tool_config=ToolConfig(
-            tool_choice="auto",
-            tool_prompt_format=ToolPromptFormat.python_list,
-            system_message_behavior=SystemMessageBehavior.replace,
-        ),
-    )
-    messages = chat_completion_request_to_messages(request, MODEL3_2)
-    assert len(messages) == 2
-    assert interleaved_content_as_str(messages[0].content).endswith(system_prompt)
-    assert "Environment: ipython" in interleaved_content_as_str(messages[0].content)
-    assert messages[-1].content == content
-
-
-async def test_replace_system_message_behavior_custom_tools():
-    content = "Hello !"
-    system_prompt = "You are a pirate"
-    request = ChatCompletionRequest(
-        model=MODEL,
-        messages=[
-            SystemMessage(content=system_prompt),
-            UserMessage(content=content),
-        ],
-        tools=[
-            ToolDefinition(tool_name=BuiltinTool.code_interpreter),
-            ToolDefinition(
-                tool_name="custom1",
-                description="custom1 tool",
-                input_schema={
-                    "type": "object",
-                    "properties": {
-                        "param1": {
-                            "type": "str",
-                            "description": "param1 description",
-                        },
-                    },
-                    "required": ["param1"],
-                },
-            ),
-        ],
-        tool_config=ToolConfig(
-            tool_choice="auto",
-            tool_prompt_format=ToolPromptFormat.python_list,
-            system_message_behavior=SystemMessageBehavior.replace,
-        ),
-    )
-    messages = chat_completion_request_to_messages(request, MODEL3_2)
-
-    assert len(messages) == 2
-    assert interleaved_content_as_str(messages[0].content).endswith(system_prompt)
-    assert "Environment: ipython" in interleaved_content_as_str(messages[0].content)
-    assert messages[-1].content == content
-
-
-async def test_replace_system_message_behavior_custom_tools_with_template():
-    content = "Hello !"
-    system_prompt = "You are a pirate {{ function_description }}"
-    request = ChatCompletionRequest(
-        model=MODEL,
-        messages=[
-            SystemMessage(content=system_prompt),
-            UserMessage(content=content),
-        ],
-        tools=[
-            ToolDefinition(tool_name=BuiltinTool.code_interpreter),
-            ToolDefinition(
-                tool_name="custom1",
-                description="custom1 tool",
-                input_schema={
-                    "type": "object",
-                    "properties": {
-                        "param1": {
-                            "type": "str",
-                            "description": "param1 description",
-                        },
-                    },
-                    "required": ["param1"],
-                },
-            ),
-        ],
-        tool_config=ToolConfig(
-            tool_choice="auto",
-            tool_prompt_format=ToolPromptFormat.python_list,
-            system_message_behavior=SystemMessageBehavior.replace,
-        ),
-    )
-    messages = chat_completion_request_to_messages(request, MODEL3_2)
-
-    assert len(messages) == 2
-    assert "Environment: ipython" in interleaved_content_as_str(messages[0].content)
-    assert "You are a pirate" in interleaved_content_as_str(messages[0].content)
-    # function description is present in the system prompt
-    assert '"name": "custom1"' in interleaved_content_as_str(messages[0].content)
-    assert messages[-1].content == content

tests/unit/prompts/prompts/conftest.py

@@ -18,7 +18,7 @@ from llama_stack.core.storage.datatypes import (
     SqlStoreReference,
     StorageConfig,
 )
-from llama_stack.providers.utils.kvstore import register_kvstore_backends
+from llama_stack.core.storage.kvstore import register_kvstore_backends
 
 
 @pytest.fixture

tests/unit/providers/agents/meta_reference/fixtures/__init__.py

@@ -0,0 +1,23 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+import os
+
+import yaml
+
+from llama_stack_api.inference import (
+    OpenAIChatCompletion,
+)
+
+FIXTURES_DIR = os.path.dirname(os.path.abspath(__file__))
+
+
+def load_chat_completion_fixture(filename: str) -> OpenAIChatCompletion:
+    fixture_path = os.path.join(FIXTURES_DIR, filename)
+
+    with open(fixture_path) as f:
+        data = yaml.safe_load(f)
+    return OpenAIChatCompletion(**data)

tests/unit/providers/agents/meta_reference/fixtures/simple_chat_completion.yaml

@@ -0,0 +1,9 @@
+id: chat-completion-123
+choices:
+  - message:
+      content: "Dublin"
+      role: assistant
+    finish_reason: stop
+    index: 0
+created: 1234567890
+model: meta-llama/Llama-3.1-8B-Instruct

tests/unit/providers/agents/meta_reference/fixtures/tool_call_completion.yaml

@@ -0,0 +1,14 @@
+id: chat-completion-123
+choices:
+  - message:
+      tool_calls:
+        - id: tool_call_123
+          type: function
+          function:
+            name: web_search
+            arguments: '{"query":"What is the capital of Ireland?"}'
+      role: assistant
+    finish_reason: stop
+    index: 0
+created: 1234567890
+model: meta-llama/Llama-3.1-8B-Instruct

tests/unit/providers/agents/meta_reference/test_openai_responses_conversations.py

@@ -0,0 +1,249 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+
+import pytest
+
+from llama_stack_api.common.errors import (
+    ConversationNotFoundError,
+    InvalidConversationIdError,
+)
+from llama_stack_api.conversations import (
+    ConversationItemList,
+)
+from llama_stack_api.openai_responses import (
+    OpenAIResponseMessage,
+    OpenAIResponseObject,
+    OpenAIResponseObjectStreamResponseCompleted,
+    OpenAIResponseObjectStreamResponseOutputItemDone,
+    OpenAIResponseOutputMessageContentOutputText,
+)
+
+# Import existing fixtures from the main responses test file
+pytest_plugins = ["tests.unit.providers.agents.meta_reference.test_openai_responses"]
+
+from llama_stack.providers.inline.agents.meta_reference.responses.openai_responses import (
+    OpenAIResponsesImpl,
+)
+
+
+@pytest.fixture
+def responses_impl_with_conversations(
+    mock_inference_api,
+    mock_tool_groups_api,
+    mock_tool_runtime_api,
+    mock_responses_store,
+    mock_vector_io_api,
+    mock_conversations_api,
+    mock_safety_api,
+):
+    """Create OpenAIResponsesImpl instance with conversations API."""
+    return OpenAIResponsesImpl(
+        inference_api=mock_inference_api,
+        tool_groups_api=mock_tool_groups_api,
+        tool_runtime_api=mock_tool_runtime_api,
+        responses_store=mock_responses_store,
+        vector_io_api=mock_vector_io_api,
+        conversations_api=mock_conversations_api,
+        safety_api=mock_safety_api,
+    )
+
+
+class TestConversationValidation:
+    """Test conversation ID validation logic."""
+
+    async def test_nonexistent_conversation_raises_error(
+        self, responses_impl_with_conversations, mock_conversations_api
+    ):
+        """Test that ConversationNotFoundError is raised for non-existent conversation."""
+        conv_id = "conv_nonexistent"
+
+        # Mock conversation not found
+        mock_conversations_api.list_items.side_effect = ConversationNotFoundError("conv_nonexistent")
+
+        with pytest.raises(ConversationNotFoundError):
+            await responses_impl_with_conversations.create_openai_response(
+                input="Hello", model="test-model", conversation=conv_id, stream=False
+            )
+
+
+class TestMessageSyncing:
+    """Test message syncing to conversations."""
+
+    async def test_sync_response_to_conversation_simple(
+        self, responses_impl_with_conversations, mock_conversations_api
+    ):
+        """Test syncing simple response to conversation."""
+        conv_id = "conv_test123"
+        input_text = "What are the 5 Ds of dodgeball?"
+
+        # Output items (what the model generated)
+        output_items = [
+            OpenAIResponseMessage(
+                id="msg_response",
+                content=[
+                    OpenAIResponseOutputMessageContentOutputText(
+                        text="The 5 Ds are: Dodge, Duck, Dip, Dive, and Dodge.", type="output_text", annotations=[]
+                    )
+                ],
+                role="assistant",
+                status="completed",
+                type="message",
+            )
+        ]
+
+        await responses_impl_with_conversations._sync_response_to_conversation(conv_id, input_text, output_items)
+
+        # should call add_items with user input and assistant response
+        mock_conversations_api.add_items.assert_called_once()
+        call_args = mock_conversations_api.add_items.call_args
+
+        assert call_args[0][0] == conv_id  # conversation_id
+        items = call_args[0][1]  # conversation_items
+
+        assert len(items) == 2
+        # User message
+        assert items[0].type == "message"
+        assert items[0].role == "user"
+        assert items[0].content[0].type == "input_text"
+        assert items[0].content[0].text == input_text
+
+        # Assistant message
+        assert items[1].type == "message"
+        assert items[1].role == "assistant"
+
+    async def test_sync_response_to_conversation_api_error(
+        self, responses_impl_with_conversations, mock_conversations_api
+    ):
+        mock_conversations_api.add_items.side_effect = Exception("API Error")
+        output_items = []
+
+        # matching the behavior of OpenAI here
+        with pytest.raises(Exception, match="API Error"):
+            await responses_impl_with_conversations._sync_response_to_conversation(
+                "conv_test123", "Hello", output_items
+            )
+
+    async def test_sync_with_list_input(self, responses_impl_with_conversations, mock_conversations_api):
+        """Test syncing with list of input messages."""
+        conv_id = "conv_test123"
+        input_messages = [
+            OpenAIResponseMessage(role="user", content=[{"type": "input_text", "text": "First message"}]),
+        ]
+        output_items = [
+            OpenAIResponseMessage(
+                id="msg_response",
+                content=[OpenAIResponseOutputMessageContentOutputText(text="Response", type="output_text")],
+                role="assistant",
+                status="completed",
+                type="message",
+            )
+        ]
+
+        await responses_impl_with_conversations._sync_response_to_conversation(conv_id, input_messages, output_items)
+
+        mock_conversations_api.add_items.assert_called_once()
+        call_args = mock_conversations_api.add_items.call_args
+
+        items = call_args[0][1]
+        # Should have input message + output message
+        assert len(items) == 2
+
+
+class TestIntegrationWorkflow:
+    """Integration tests for the full conversation workflow."""
+
+    async def test_create_response_with_valid_conversation(
+        self, responses_impl_with_conversations, mock_conversations_api
+    ):
+        """Test creating a response with a valid conversation parameter."""
+        mock_conversations_api.list_items.return_value = ConversationItemList(
+            data=[], first_id=None, has_more=False, last_id=None, object="list"
+        )
+
+        async def mock_streaming_response(*args, **kwargs):
+            message_item = OpenAIResponseMessage(
+                id="msg_response",
+                content=[
+                    OpenAIResponseOutputMessageContentOutputText(
+                        text="Test response", type="output_text", annotations=[]
+                    )
+                ],
+                role="assistant",
+                status="completed",
+                type="message",
+            )
+
+            # Emit output_item.done event first (needed for conversation sync)
+            yield OpenAIResponseObjectStreamResponseOutputItemDone(
+                response_id="resp_test123",
+                item=message_item,
+                output_index=0,
+                sequence_number=1,
+                type="response.output_item.done",
+            )
+
+            # Then emit response.completed
+            mock_response = OpenAIResponseObject(
+                id="resp_test123",
+                created_at=1234567890,
+                model="test-model",
+                object="response",
+                output=[message_item],
+                status="completed",
+            )
+
+            yield OpenAIResponseObjectStreamResponseCompleted(response=mock_response, type="response.completed")
+
+        responses_impl_with_conversations._create_streaming_response = mock_streaming_response
+
+        input_text = "Hello, how are you?"
+        conversation_id = "conv_test123"
+
+        response = await responses_impl_with_conversations.create_openai_response(
+            input=input_text, model="test-model", conversation=conversation_id, stream=False
+        )
+
+        assert response is not None
+        assert response.id == "resp_test123"
+
+        # Note: conversation sync happens inside _create_streaming_response,
+        # which we're mocking here, so we can't test it in this unit test.
+        # The sync logic is tested separately in TestMessageSyncing.
+
+    async def test_create_response_with_invalid_conversation_id(self, responses_impl_with_conversations):
+        """Test creating a response with an invalid conversation ID."""
+        with pytest.raises(InvalidConversationIdError) as exc_info:
+            await responses_impl_with_conversations.create_openai_response(
+                input="Hello", model="test-model", conversation="invalid_id", stream=False
+            )
+
+        assert "Expected an ID that begins with 'conv_'" in str(exc_info.value)
+
+    async def test_create_response_with_nonexistent_conversation(
+        self, responses_impl_with_conversations, mock_conversations_api
+    ):
+        """Test creating a response with a non-existent conversation."""
+        mock_conversations_api.list_items.side_effect = ConversationNotFoundError("conv_nonexistent")
+
+        with pytest.raises(ConversationNotFoundError) as exc_info:
+            await responses_impl_with_conversations.create_openai_response(
+                input="Hello", model="test-model", conversation="conv_nonexistent", stream=False
+            )
+
+        assert "not found" in str(exc_info.value)
+
+    async def test_conversation_and_previous_response_id(
+        self, responses_impl_with_conversations, mock_conversations_api, mock_responses_store
+    ):
+        with pytest.raises(ValueError) as exc_info:
+            await responses_impl_with_conversations.create_openai_response(
+                input="test", model="test", conversation="conv_123", previous_response_id="resp_123"
+            )
+
+        assert "Mutually exclusive parameters" in str(exc_info.value)
+        assert "previous_response_id" in str(exc_info.value)
+        assert "conversation" in str(exc_info.value)

tests/unit/providers/agents/meta_reference/test_response_conversion_utils.py

@@ -0,0 +1,367 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+
+import pytest
+
+from llama_stack.providers.inline.agents.meta_reference.responses.utils import (
+    _extract_citations_from_text,
+    convert_chat_choice_to_response_message,
+    convert_response_content_to_chat_content,
+    convert_response_input_to_chat_messages,
+    convert_response_text_to_chat_response_format,
+    get_message_type_by_role,
+    is_function_tool_call,
+)
+from llama_stack_api.inference import (
+    OpenAIAssistantMessageParam,
+    OpenAIChatCompletionContentPartImageParam,
+    OpenAIChatCompletionContentPartTextParam,
+    OpenAIChatCompletionToolCall,
+    OpenAIChatCompletionToolCallFunction,
+    OpenAIChoice,
+    OpenAIDeveloperMessageParam,
+    OpenAIResponseFormatJSONObject,
+    OpenAIResponseFormatJSONSchema,
+    OpenAIResponseFormatText,
+    OpenAISystemMessageParam,
+    OpenAIToolMessageParam,
+    OpenAIUserMessageParam,
+)
+from llama_stack_api.openai_responses import (
+    OpenAIResponseAnnotationFileCitation,
+    OpenAIResponseInputFunctionToolCallOutput,
+    OpenAIResponseInputMessageContentImage,
+    OpenAIResponseInputMessageContentText,
+    OpenAIResponseInputToolFunction,
+    OpenAIResponseInputToolWebSearch,
+    OpenAIResponseMessage,
+    OpenAIResponseOutputMessageContentOutputText,
+    OpenAIResponseOutputMessageFunctionToolCall,
+    OpenAIResponseText,
+    OpenAIResponseTextFormat,
+)
+
+
+class TestConvertChatChoiceToResponseMessage:
+    async def test_convert_string_content(self):
+        choice = OpenAIChoice(
+            message=OpenAIAssistantMessageParam(content="Test message"),
+            finish_reason="stop",
+            index=0,
+        )
+
+        result = await convert_chat_choice_to_response_message(choice)
+
+        assert result.role == "assistant"
+        assert result.status == "completed"
+        assert len(result.content) == 1
+        assert isinstance(result.content[0], OpenAIResponseOutputMessageContentOutputText)
+        assert result.content[0].text == "Test message"
+
+    async def test_convert_text_param_content(self):
+        choice = OpenAIChoice(
+            message=OpenAIAssistantMessageParam(
+                content=[OpenAIChatCompletionContentPartTextParam(text="Test text param")]
+            ),
+            finish_reason="stop",
+            index=0,
+        )
+
+        with pytest.raises(ValueError) as exc_info:
+            await convert_chat_choice_to_response_message(choice)
+
+        assert "does not yet support output content type" in str(exc_info.value)
+
+
+class TestConvertResponseContentToChatContent:
+    async def test_convert_string_content(self):
+        result = await convert_response_content_to_chat_content("Simple string")
+        assert result == "Simple string"
+
+    async def test_convert_text_content_parts(self):
+        content = [
+            OpenAIResponseInputMessageContentText(text="First part"),
+            OpenAIResponseOutputMessageContentOutputText(text="Second part"),
+        ]
+
+        result = await convert_response_content_to_chat_content(content)
+
+        assert len(result) == 2
+        assert isinstance(result[0], OpenAIChatCompletionContentPartTextParam)
+        assert result[0].text == "First part"
+        assert isinstance(result[1], OpenAIChatCompletionContentPartTextParam)
+        assert result[1].text == "Second part"
+
+    async def test_convert_image_content(self):
+        content = [OpenAIResponseInputMessageContentImage(image_url="https://example.com/image.jpg", detail="high")]
+
+        result = await convert_response_content_to_chat_content(content)
+
+        assert len(result) == 1
+        assert isinstance(result[0], OpenAIChatCompletionContentPartImageParam)
+        assert result[0].image_url.url == "https://example.com/image.jpg"
+        assert result[0].image_url.detail == "high"
+
+
+class TestConvertResponseInputToChatMessages:
+    async def test_convert_string_input(self):
+        result = await convert_response_input_to_chat_messages("User message")
+
+        assert len(result) == 1
+        assert isinstance(result[0], OpenAIUserMessageParam)
+        assert result[0].content == "User message"
+
+    async def test_convert_function_tool_call_output(self):
+        input_items = [
+            OpenAIResponseOutputMessageFunctionToolCall(
+                call_id="call_123",
+                name="test_function",
+                arguments='{"param": "value"}',
+            ),
+            OpenAIResponseInputFunctionToolCallOutput(
+                output="Tool output",
+                call_id="call_123",
+            ),
+        ]
+
+        result = await convert_response_input_to_chat_messages(input_items)
+
+        assert len(result) == 2
+        assert isinstance(result[0], OpenAIAssistantMessageParam)
+        assert result[0].tool_calls[0].id == "call_123"
+        assert result[0].tool_calls[0].function.name == "test_function"
+        assert result[0].tool_calls[0].function.arguments == '{"param": "value"}'
+        assert isinstance(result[1], OpenAIToolMessageParam)
+        assert result[1].content == "Tool output"
+        assert result[1].tool_call_id == "call_123"
+
+    async def test_convert_function_tool_call(self):
+        input_items = [
+            OpenAIResponseOutputMessageFunctionToolCall(
+                call_id="call_456",
+                name="test_function",
+                arguments='{"param": "value"}',
+            )
+        ]
+
+        result = await convert_response_input_to_chat_messages(input_items)
+
+        assert len(result) == 1
+        assert isinstance(result[0], OpenAIAssistantMessageParam)
+        assert len(result[0].tool_calls) == 1
+        assert result[0].tool_calls[0].id == "call_456"
+        assert result[0].tool_calls[0].function.name == "test_function"
+        assert result[0].tool_calls[0].function.arguments == '{"param": "value"}'
+
+    async def test_convert_function_call_ordering(self):
+        input_items = [
+            OpenAIResponseOutputMessageFunctionToolCall(
+                call_id="call_123",
+                name="test_function_a",
+                arguments='{"param": "value"}',
+            ),
+            OpenAIResponseOutputMessageFunctionToolCall(
+                call_id="call_456",
+                name="test_function_b",
+                arguments='{"param": "value"}',
+            ),
+            OpenAIResponseInputFunctionToolCallOutput(
+                output="AAA",
+                call_id="call_123",
+            ),
+            OpenAIResponseInputFunctionToolCallOutput(
+                output="BBB",
+                call_id="call_456",
+            ),
+        ]
+
+        result = await convert_response_input_to_chat_messages(input_items)
+        assert len(result) == 4
+        assert isinstance(result[0], OpenAIAssistantMessageParam)
+        assert len(result[0].tool_calls) == 1
+        assert result[0].tool_calls[0].id == "call_123"
+        assert result[0].tool_calls[0].function.name == "test_function_a"
+        assert result[0].tool_calls[0].function.arguments == '{"param": "value"}'
+        assert isinstance(result[1], OpenAIToolMessageParam)
+        assert result[1].content == "AAA"
+        assert result[1].tool_call_id == "call_123"
+        assert isinstance(result[2], OpenAIAssistantMessageParam)
+        assert len(result[2].tool_calls) == 1
+        assert result[2].tool_calls[0].id == "call_456"
+        assert result[2].tool_calls[0].function.name == "test_function_b"
+        assert result[2].tool_calls[0].function.arguments == '{"param": "value"}'
+        assert isinstance(result[3], OpenAIToolMessageParam)
+        assert result[3].content == "BBB"
+        assert result[3].tool_call_id == "call_456"
+
+    async def test_convert_response_message(self):
+        input_items = [
+            OpenAIResponseMessage(
+                role="user",
+                content=[OpenAIResponseInputMessageContentText(text="User text")],
+            )
+        ]
+
+        result = await convert_response_input_to_chat_messages(input_items)
+
+        assert len(result) == 1
+        assert isinstance(result[0], OpenAIUserMessageParam)
+        # Content should be converted to chat content format
+        assert len(result[0].content) == 1
+        assert result[0].content[0].text == "User text"
+
+
+class TestConvertResponseTextToChatResponseFormat:
+    async def test_convert_text_format(self):
+        text = OpenAIResponseText(format=OpenAIResponseTextFormat(type="text"))
+        result = await convert_response_text_to_chat_response_format(text)
+
+        assert isinstance(result, OpenAIResponseFormatText)
+        assert result.type == "text"
+
+    async def test_convert_json_object_format(self):
+        text = OpenAIResponseText(format={"type": "json_object"})
+        result = await convert_response_text_to_chat_response_format(text)
+
+        assert isinstance(result, OpenAIResponseFormatJSONObject)
+
+    async def test_convert_json_schema_format(self):
+        schema_def = {"type": "object", "properties": {"test": {"type": "string"}}}
+        text = OpenAIResponseText(
+            format={
+                "type": "json_schema",
+                "name": "test_schema",
+                "schema": schema_def,
+            }
+        )
+        result = await convert_response_text_to_chat_response_format(text)
+
+        assert isinstance(result, OpenAIResponseFormatJSONSchema)
+        assert result.json_schema["name"] == "test_schema"
+        assert result.json_schema["schema"] == schema_def
+
+    async def test_default_text_format(self):
+        text = OpenAIResponseText()
+        result = await convert_response_text_to_chat_response_format(text)
+
+        assert isinstance(result, OpenAIResponseFormatText)
+        assert result.type == "text"
+
+
+class TestGetMessageTypeByRole:
+    async def test_user_role(self):
+        result = await get_message_type_by_role("user")
+        assert result == OpenAIUserMessageParam
+
+    async def test_system_role(self):
+        result = await get_message_type_by_role("system")
+        assert result == OpenAISystemMessageParam
+
+    async def test_assistant_role(self):
+        result = await get_message_type_by_role("assistant")
+        assert result == OpenAIAssistantMessageParam
+
+    async def test_developer_role(self):
+        result = await get_message_type_by_role("developer")
+        assert result == OpenAIDeveloperMessageParam
+
+    async def test_unknown_role(self):
+        result = await get_message_type_by_role("unknown")
+        assert result is None
+
+
+class TestIsFunctionToolCall:
+    def test_is_function_tool_call_true(self):
+        tool_call = OpenAIChatCompletionToolCall(
+            index=0,
+            id="call_123",
+            function=OpenAIChatCompletionToolCallFunction(
+                name="test_function",
+                arguments="{}",
+            ),
+        )
+        tools = [
+            OpenAIResponseInputToolFunction(
+                type="function", name="test_function", parameters={"type": "object", "properties": {}}
+            ),
+            OpenAIResponseInputToolWebSearch(type="web_search"),
+        ]
+
+        result = is_function_tool_call(tool_call, tools)
+        assert result is True
+
+    def test_is_function_tool_call_false_different_name(self):
+        tool_call = OpenAIChatCompletionToolCall(
+            index=0,
+            id="call_123",
+            function=OpenAIChatCompletionToolCallFunction(
+                name="other_function",
+                arguments="{}",
+            ),
+        )
+        tools = [
+            OpenAIResponseInputToolFunction(
+                type="function", name="test_function", parameters={"type": "object", "properties": {}}
+            ),
+        ]
+
+        result = is_function_tool_call(tool_call, tools)
+        assert result is False
+
+    def test_is_function_tool_call_false_no_function(self):
+        tool_call = OpenAIChatCompletionToolCall(
+            index=0,
+            id="call_123",
+            function=None,
+        )
+        tools = [
+            OpenAIResponseInputToolFunction(
+                type="function", name="test_function", parameters={"type": "object", "properties": {}}
+            ),
+        ]
+
+        result = is_function_tool_call(tool_call, tools)
+        assert result is False
+
+    def test_is_function_tool_call_false_wrong_type(self):
+        tool_call = OpenAIChatCompletionToolCall(
+            index=0,
+            id="call_123",
+            function=OpenAIChatCompletionToolCallFunction(
+                name="web_search",
+                arguments="{}",
+            ),
+        )
+        tools = [
+            OpenAIResponseInputToolWebSearch(type="web_search"),
+        ]
+
+        result = is_function_tool_call(tool_call, tools)
+        assert result is False
+
+
+class TestExtractCitationsFromText:
+    def test_extract_citations_and_annotations(self):
+        text = "Start [not-a-file]. New source <|file-abc123|>. "
+        text += "Other source <|file-def456|>? Repeat source <|file-abc123|>! No citation."
+        file_mapping = {"file-abc123": "doc1.pdf", "file-def456": "doc2.txt"}
+
+        annotations, cleaned_text = _extract_citations_from_text(text, file_mapping)
+
+        expected_annotations = [
+            OpenAIResponseAnnotationFileCitation(file_id="file-abc123", filename="doc1.pdf", index=30),
+            OpenAIResponseAnnotationFileCitation(file_id="file-def456", filename="doc2.txt", index=44),
+            OpenAIResponseAnnotationFileCitation(file_id="file-abc123", filename="doc1.pdf", index=59),
+        ]
+        expected_clean_text = "Start [not-a-file]. New source. Other source? Repeat source! No citation."
+
+        assert cleaned_text == expected_clean_text
+        assert annotations == expected_annotations
+        # OpenAI cites at the end of the sentence
+        assert cleaned_text[expected_annotations[0].index] == "."
+        assert cleaned_text[expected_annotations[1].index] == "?"
+        assert cleaned_text[expected_annotations[2].index] == "!"

tests/unit/providers/agents/meta_reference/test_response_tool_context.py

@@ -0,0 +1,183 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+
+from llama_stack.providers.inline.agents.meta_reference.responses.types import ToolContext
+from llama_stack_api.openai_responses import (
+    MCPListToolsTool,
+    OpenAIResponseInputToolFileSearch,
+    OpenAIResponseInputToolFunction,
+    OpenAIResponseInputToolMCP,
+    OpenAIResponseInputToolWebSearch,
+    OpenAIResponseObject,
+    OpenAIResponseOutputMessageMCPListTools,
+    OpenAIResponseToolMCP,
+)
+
+
+class TestToolContext:
+    def test_no_tools(self):
+        tools = []
+        context = ToolContext(tools)
+        previous_response = OpenAIResponseObject(created_at=1234, id="test", model="mymodel", output=[], status="")
+        context.recover_tools_from_previous_response(previous_response)
+
+        assert len(context.tools_to_process) == 0
+        assert len(context.previous_tools) == 0
+        assert len(context.previous_tool_listings) == 0
+
+    def test_no_previous_tools(self):
+        tools = [
+            OpenAIResponseInputToolFileSearch(vector_store_ids=["fake"]),
+            OpenAIResponseInputToolMCP(server_label="label", server_url="url"),
+        ]
+        context = ToolContext(tools)
+        previous_response = OpenAIResponseObject(created_at=1234, id="test", model="mymodel", output=[], status="")
+        context.recover_tools_from_previous_response(previous_response)
+
+        assert len(context.tools_to_process) == 2
+        assert len(context.previous_tools) == 0
+        assert len(context.previous_tool_listings) == 0
+
+    def test_reusable_server(self):
+        tools = [
+            OpenAIResponseInputToolFileSearch(vector_store_ids=["fake"]),
+            OpenAIResponseInputToolMCP(server_label="alabel", server_url="aurl"),
+        ]
+        context = ToolContext(tools)
+        output = [
+            OpenAIResponseOutputMessageMCPListTools(
+                id="test", server_label="alabel", tools=[MCPListToolsTool(name="test_tool", input_schema={})]
+            )
+        ]
+        previous_response = OpenAIResponseObject(created_at=1234, id="test", model="fake", output=output, status="")
+        previous_response.tools = [
+            OpenAIResponseInputToolFileSearch(vector_store_ids=["fake"]),
+            OpenAIResponseToolMCP(server_label="alabel"),
+        ]
+        context.recover_tools_from_previous_response(previous_response)
+
+        assert len(context.tools_to_process) == 1
+        assert context.tools_to_process[0].type == "file_search"
+        assert len(context.previous_tools) == 1
+        assert context.previous_tools["test_tool"].server_label == "alabel"
+        assert context.previous_tools["test_tool"].server_url == "aurl"
+        assert len(context.previous_tool_listings) == 1
+        assert len(context.previous_tool_listings[0].tools) == 1
+        assert context.previous_tool_listings[0].server_label == "alabel"
+
+    def test_multiple_reusable_servers(self):
+        tools = [
+            OpenAIResponseInputToolFunction(name="fake", parameters=None),
+            OpenAIResponseInputToolMCP(server_label="anotherlabel", server_url="anotherurl"),
+            OpenAIResponseInputToolWebSearch(),
+            OpenAIResponseInputToolMCP(server_label="alabel", server_url="aurl"),
+        ]
+        context = ToolContext(tools)
+        output = [
+            OpenAIResponseOutputMessageMCPListTools(
+                id="test1", server_label="alabel", tools=[MCPListToolsTool(name="test_tool", input_schema={})]
+            ),
+            OpenAIResponseOutputMessageMCPListTools(
+                id="test2",
+                server_label="anotherlabel",
+                tools=[MCPListToolsTool(name="some_other_tool", input_schema={})],
+            ),
+        ]
+        previous_response = OpenAIResponseObject(created_at=1234, id="test", model="fake", output=output, status="")
+        previous_response.tools = [
+            OpenAIResponseInputToolFunction(name="fake", parameters=None),
+            OpenAIResponseToolMCP(server_label="anotherlabel", server_url="anotherurl"),
+            OpenAIResponseInputToolWebSearch(type="web_search"),
+            OpenAIResponseToolMCP(server_label="alabel", server_url="aurl"),
+        ]
+        context.recover_tools_from_previous_response(previous_response)
+
+        assert len(context.tools_to_process) == 2
+        assert context.tools_to_process[0].type == "function"
+        assert context.tools_to_process[1].type == "web_search"
+        assert len(context.previous_tools) == 2
+        assert context.previous_tools["test_tool"].server_label == "alabel"
+        assert context.previous_tools["test_tool"].server_url == "aurl"
+        assert context.previous_tools["some_other_tool"].server_label == "anotherlabel"
+        assert context.previous_tools["some_other_tool"].server_url == "anotherurl"
+        assert len(context.previous_tool_listings) == 2
+        assert len(context.previous_tool_listings[0].tools) == 1
+        assert context.previous_tool_listings[0].server_label == "alabel"
+        assert len(context.previous_tool_listings[1].tools) == 1
+        assert context.previous_tool_listings[1].server_label == "anotherlabel"
+
+    def test_multiple_servers_only_one_reusable(self):
+        tools = [
+            OpenAIResponseInputToolFunction(name="fake", parameters=None),
+            OpenAIResponseInputToolMCP(server_label="anotherlabel", server_url="anotherurl"),
+            OpenAIResponseInputToolWebSearch(type="web_search"),
+            OpenAIResponseInputToolMCP(server_label="alabel", server_url="aurl"),
+        ]
+        context = ToolContext(tools)
+        output = [
+            OpenAIResponseOutputMessageMCPListTools(
+                id="test2",
+                server_label="anotherlabel",
+                tools=[MCPListToolsTool(name="some_other_tool", input_schema={})],
+            )
+        ]
+        previous_response = OpenAIResponseObject(created_at=1234, id="test", model="fake", output=output, status="")
+        previous_response.tools = [
+            OpenAIResponseInputToolFunction(name="fake", parameters=None),
+            OpenAIResponseToolMCP(server_label="anotherlabel", server_url="anotherurl"),
+            OpenAIResponseInputToolWebSearch(type="web_search"),
+        ]
+        context.recover_tools_from_previous_response(previous_response)
+
+        assert len(context.tools_to_process) == 3
+        assert context.tools_to_process[0].type == "function"
+        assert context.tools_to_process[1].type == "web_search"
+        assert context.tools_to_process[2].type == "mcp"
+        assert len(context.previous_tools) == 1
+        assert context.previous_tools["some_other_tool"].server_label == "anotherlabel"
+        assert context.previous_tools["some_other_tool"].server_url == "anotherurl"
+        assert len(context.previous_tool_listings) == 1
+        assert len(context.previous_tool_listings[0].tools) == 1
+        assert context.previous_tool_listings[0].server_label == "anotherlabel"
+
+    def test_mismatched_allowed_tools(self):
+        tools = [
+            OpenAIResponseInputToolFunction(name="fake", parameters=None),
+            OpenAIResponseInputToolMCP(server_label="anotherlabel", server_url="anotherurl"),
+            OpenAIResponseInputToolWebSearch(type="web_search"),
+            OpenAIResponseInputToolMCP(server_label="alabel", server_url="aurl", allowed_tools=["test_tool_2"]),
+        ]
+        context = ToolContext(tools)
+        output = [
+            OpenAIResponseOutputMessageMCPListTools(
+                id="test1", server_label="alabel", tools=[MCPListToolsTool(name="test_tool_1", input_schema={})]
+            ),
+            OpenAIResponseOutputMessageMCPListTools(
+                id="test2",
+                server_label="anotherlabel",
+                tools=[MCPListToolsTool(name="some_other_tool", input_schema={})],
+            ),
+        ]
+        previous_response = OpenAIResponseObject(created_at=1234, id="test", model="fake", output=output, status="")
+        previous_response.tools = [
+            OpenAIResponseInputToolFunction(name="fake", parameters=None),
+            OpenAIResponseToolMCP(server_label="anotherlabel", server_url="anotherurl"),
+            OpenAIResponseInputToolWebSearch(type="web_search"),
+            OpenAIResponseToolMCP(server_label="alabel", server_url="aurl"),
+        ]
+        context.recover_tools_from_previous_response(previous_response)
+
+        assert len(context.tools_to_process) == 3
+        assert context.tools_to_process[0].type == "function"
+        assert context.tools_to_process[1].type == "web_search"
+        assert context.tools_to_process[2].type == "mcp"
+        assert len(context.previous_tools) == 1
+        assert context.previous_tools["some_other_tool"].server_label == "anotherlabel"
+        assert context.previous_tools["some_other_tool"].server_url == "anotherurl"
+        assert len(context.previous_tool_listings) == 1
+        assert len(context.previous_tool_listings[0].tools) == 1
+        assert context.previous_tool_listings[0].server_label == "anotherlabel"

tests/unit/providers/agents/meta_reference/test_responses_safety_utils.py

@@ -0,0 +1,155 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+from unittest.mock import AsyncMock
+
+import pytest
+
+from llama_stack.providers.inline.agents.meta_reference.responses.openai_responses import (
+    OpenAIResponsesImpl,
+)
+from llama_stack.providers.inline.agents.meta_reference.responses.utils import (
+    extract_guardrail_ids,
+    run_guardrails,
+)
+from llama_stack_api.agents import ResponseGuardrailSpec
+from llama_stack_api.safety import ModerationObject, ModerationObjectResults
+
+
+@pytest.fixture
+def mock_apis():
+    """Create mock APIs for testing."""
+    return {
+        "inference_api": AsyncMock(),
+        "tool_groups_api": AsyncMock(),
+        "tool_runtime_api": AsyncMock(),
+        "responses_store": AsyncMock(),
+        "vector_io_api": AsyncMock(),
+        "conversations_api": AsyncMock(),
+        "safety_api": AsyncMock(),
+    }
+
+
+@pytest.fixture
+def responses_impl(mock_apis):
+    """Create OpenAIResponsesImpl instance with mocked dependencies."""
+    return OpenAIResponsesImpl(**mock_apis)
+
+
+def test_extract_guardrail_ids_from_strings(responses_impl):
+    """Test extraction from simple string guardrail IDs."""
+    guardrails = ["llama-guard", "content-filter", "nsfw-detector"]
+    result = extract_guardrail_ids(guardrails)
+    assert result == ["llama-guard", "content-filter", "nsfw-detector"]
+
+
+def test_extract_guardrail_ids_from_objects(responses_impl):
+    """Test extraction from ResponseGuardrailSpec objects."""
+    guardrails = [
+        ResponseGuardrailSpec(type="llama-guard"),
+        ResponseGuardrailSpec(type="content-filter"),
+    ]
+    result = extract_guardrail_ids(guardrails)
+    assert result == ["llama-guard", "content-filter"]
+
+
+def test_extract_guardrail_ids_mixed_formats(responses_impl):
+    """Test extraction from mixed string and object formats."""
+    guardrails = [
+        "llama-guard",
+        ResponseGuardrailSpec(type="content-filter"),
+        "nsfw-detector",
+    ]
+    result = extract_guardrail_ids(guardrails)
+    assert result == ["llama-guard", "content-filter", "nsfw-detector"]
+
+
+def test_extract_guardrail_ids_none_input(responses_impl):
+    """Test extraction with None input."""
+    result = extract_guardrail_ids(None)
+    assert result == []
+
+
+def test_extract_guardrail_ids_empty_list(responses_impl):
+    """Test extraction with empty list."""
+    result = extract_guardrail_ids([])
+    assert result == []
+
+
+def test_extract_guardrail_ids_unknown_format(responses_impl):
+    """Test extraction with unknown guardrail format raises ValueError."""
+    # Create an object that's neither string nor ResponseGuardrailSpec
+    unknown_object = {"invalid": "format"}  # Plain dict, not ResponseGuardrailSpec
+    guardrails = ["valid-guardrail", unknown_object, "another-guardrail"]
+    with pytest.raises(ValueError, match="Unknown guardrail format.*expected str or ResponseGuardrailSpec"):
+        extract_guardrail_ids(guardrails)
+
+
+@pytest.fixture
+def mock_safety_api():
+    """Create mock safety API for guardrails testing."""
+    safety_api = AsyncMock()
+    # Mock the routing table and shields list for guardrails lookup
+    safety_api.routing_table = AsyncMock()
+    shield = AsyncMock()
+    shield.identifier = "llama-guard"
+    shield.provider_resource_id = "llama-guard-model"
+    safety_api.routing_table.list_shields.return_value = AsyncMock(data=[shield])
+    return safety_api
+
+
+async def test_run_guardrails_no_violation(mock_safety_api):
+    """Test guardrails validation with no violations."""
+    text = "Hello world"
+    guardrail_ids = ["llama-guard"]
+
+    # Mock moderation to return non-flagged content
+    unflagged_result = ModerationObjectResults(flagged=False, categories={"violence": False})
+    mock_moderation_object = ModerationObject(id="test-mod-id", model="llama-guard-model", results=[unflagged_result])
+    mock_safety_api.run_moderation.return_value = mock_moderation_object
+
+    result = await run_guardrails(mock_safety_api, text, guardrail_ids)
+
+    assert result is None
+    # Verify run_moderation was called with the correct model
+    mock_safety_api.run_moderation.assert_called_once()
+    call_args = mock_safety_api.run_moderation.call_args
+    assert call_args[1]["model"] == "llama-guard-model"
+
+
+async def test_run_guardrails_with_violation(mock_safety_api):
+    """Test guardrails validation with safety violation."""
+    text = "Harmful content"
+    guardrail_ids = ["llama-guard"]
+
+    # Mock moderation to return flagged content
+    flagged_result = ModerationObjectResults(
+        flagged=True,
+        categories={"violence": True},
+        user_message="Content flagged by moderation",
+        metadata={"violation_type": ["S1"]},
+    )
+    mock_moderation_object = ModerationObject(id="test-mod-id", model="llama-guard-model", results=[flagged_result])
+    mock_safety_api.run_moderation.return_value = mock_moderation_object
+
+    result = await run_guardrails(mock_safety_api, text, guardrail_ids)
+
+    assert result == "Content flagged by moderation (flagged for: violence) (violation type: S1)"
+
+
+async def test_run_guardrails_empty_inputs(mock_safety_api):
+    """Test guardrails validation with empty inputs."""
+    # Test empty guardrail_ids
+    result = await run_guardrails(mock_safety_api, "test", [])
+    assert result is None
+
+    # Test empty text
+    result = await run_guardrails(mock_safety_api, "", ["llama-guard"])
+    assert result is None
+
+    # Test both empty
+    result = await run_guardrails(mock_safety_api, "", [])
+    assert result is None

tests/unit/providers/batches/conftest.py

@@ -13,9 +13,9 @@ from unittest.mock import AsyncMock
 import pytest
 
 from llama_stack.core.storage.datatypes import KVStoreReference, SqliteKVStoreConfig
+from llama_stack.core.storage.kvstore import kvstore_impl, register_kvstore_backends
 from llama_stack.providers.inline.batches.reference.batches import ReferenceBatchesImpl
 from llama_stack.providers.inline.batches.reference.config import ReferenceBatchesImplConfig
-from llama_stack.providers.utils.kvstore import kvstore_impl, register_kvstore_backends
 
 
 @pytest.fixture

tests/unit/providers/batches/test_reference.py

@@ -59,8 +59,7 @@ from unittest.mock import AsyncMock, MagicMock
 
 import pytest
 
-from llama_stack.apis.batches import BatchObject
-from llama_stack.apis.common.errors import ConflictError, ResourceNotFoundError
+from llama_stack_api import BatchObject, ConflictError, ResourceNotFoundError
 
 
 class TestReferenceBatchesImpl:

tests/unit/providers/batches/test_reference_idempotency.py

@@ -44,7 +44,7 @@ import asyncio
 
 import pytest
 
-from llama_stack.apis.common.errors import ConflictError
+from llama_stack_api import ConflictError
 
 
 class TestReferenceBatchesIdempotency:

tests/unit/providers/files/conftest.py

@@ -9,8 +9,8 @@ import pytest
 from moto import mock_aws
 
 from llama_stack.core.storage.datatypes import SqliteSqlStoreConfig, SqlStoreReference
+from llama_stack.core.storage.sqlstore.sqlstore import register_sqlstore_backends
 from llama_stack.providers.remote.files.s3 import S3FilesImplConfig, get_adapter_impl
-from llama_stack.providers.utils.sqlstore.sqlstore import register_sqlstore_backends
 
 
 class MockUploadFile:

tests/unit/providers/files/test_s3_files.py

@@ -9,8 +9,7 @@ from unittest.mock import patch
 import pytest
 from botocore.exceptions import ClientError
 
-from llama_stack.apis.common.errors import ResourceNotFoundError
-from llama_stack.apis.files import OpenAIFilePurpose
+from llama_stack_api import OpenAIFilePurpose, ResourceNotFoundError
 
 
 class TestS3FilesImpl:
@@ -228,7 +227,7 @@ class TestS3FilesImpl:
 
             mock_now.return_value = 0
 
-            from llama_stack.apis.files import ExpiresAfter
+            from llama_stack_api import ExpiresAfter
 
             sample_text_file.filename = "test_expired_file"
             uploaded = await s3_provider.openai_upload_file(
@@ -260,7 +259,7 @@ class TestS3FilesImpl:
 
     async def test_unsupported_expires_after_anchor(self, s3_provider, sample_text_file):
         """Unsupported anchor value should raise ValueError."""
-        from llama_stack.apis.files import ExpiresAfter
+        from llama_stack_api import ExpiresAfter
 
         sample_text_file.filename = "test_unsupported_expires_after_anchor"
 
@@ -273,7 +272,7 @@ class TestS3FilesImpl:
 
     async def test_nonint_expires_after_seconds(self, s3_provider, sample_text_file):
         """Non-integer seconds in expires_after should raise ValueError."""
-        from llama_stack.apis.files import ExpiresAfter
+        from llama_stack_api import ExpiresAfter
 
         sample_text_file.filename = "test_nonint_expires_after_seconds"
 
@@ -286,7 +285,7 @@ class TestS3FilesImpl:
 
     async def test_expires_after_seconds_out_of_bounds(self, s3_provider, sample_text_file):
         """Seconds outside allowed range should raise ValueError."""
-        from llama_stack.apis.files import ExpiresAfter
+        from llama_stack_api import ExpiresAfter
 
         with pytest.raises(ValueError, match="greater than or equal to 3600"):
             await s3_provider.openai_upload_file(

tests/unit/providers/files/test_s3_files_auth.py

@@ -8,10 +8,9 @@ from unittest.mock import patch
 
 import pytest
 
-from llama_stack.apis.common.errors import ResourceNotFoundError
-from llama_stack.apis.files import OpenAIFilePurpose
 from llama_stack.core.datatypes import User
 from llama_stack.providers.remote.files.s3.files import S3FilesImpl
+from llama_stack_api import OpenAIFilePurpose, ResourceNotFoundError
 
 
 async def test_listing_hides_other_users_file(s3_provider, sample_text_file):
@@ -19,11 +18,11 @@ async def test_listing_hides_other_users_file(s3_provider, sample_text_file):
     user_a = User("user-a", {"roles": ["team-a"]})
     user_b = User("user-b", {"roles": ["team-b"]})
 
-    with patch("llama_stack.providers.utils.sqlstore.authorized_sqlstore.get_authenticated_user") as mock_get_user:
+    with patch("llama_stack.core.storage.sqlstore.authorized_sqlstore.get_authenticated_user") as mock_get_user:
         mock_get_user.return_value = user_a
         uploaded = await s3_provider.openai_upload_file(file=sample_text_file, purpose=OpenAIFilePurpose.ASSISTANTS)
 
-    with patch("llama_stack.providers.utils.sqlstore.authorized_sqlstore.get_authenticated_user") as mock_get_user:
+    with patch("llama_stack.core.storage.sqlstore.authorized_sqlstore.get_authenticated_user") as mock_get_user:
         mock_get_user.return_value = user_b
         listed = await s3_provider.openai_list_files()
         assert all(f.id != uploaded.id for f in listed.data)
@@ -42,11 +41,11 @@ async def test_cannot_access_other_user_file(s3_provider, sample_text_file, op):
     user_a = User("user-a", {"roles": ["team-a"]})
     user_b = User("user-b", {"roles": ["team-b"]})
 
-    with patch("llama_stack.providers.utils.sqlstore.authorized_sqlstore.get_authenticated_user") as mock_get_user:
+    with patch("llama_stack.core.storage.sqlstore.authorized_sqlstore.get_authenticated_user") as mock_get_user:
         mock_get_user.return_value = user_a
         uploaded = await s3_provider.openai_upload_file(file=sample_text_file, purpose=OpenAIFilePurpose.ASSISTANTS)
 
-    with patch("llama_stack.providers.utils.sqlstore.authorized_sqlstore.get_authenticated_user") as mock_get_user:
+    with patch("llama_stack.core.storage.sqlstore.authorized_sqlstore.get_authenticated_user") as mock_get_user:
         mock_get_user.return_value = user_b
         with pytest.raises(ResourceNotFoundError):
             await op(s3_provider, uploaded.id)
@@ -57,11 +56,11 @@ async def test_shared_role_allows_listing(s3_provider, sample_text_file):
     user_a = User("user-a", {"roles": ["shared-role"]})
     user_b = User("user-b", {"roles": ["shared-role"]})
 
-    with patch("llama_stack.providers.utils.sqlstore.authorized_sqlstore.get_authenticated_user") as mock_get_user:
+    with patch("llama_stack.core.storage.sqlstore.authorized_sqlstore.get_authenticated_user") as mock_get_user:
         mock_get_user.return_value = user_a
         uploaded = await s3_provider.openai_upload_file(file=sample_text_file, purpose=OpenAIFilePurpose.ASSISTANTS)
 
-    with patch("llama_stack.providers.utils.sqlstore.authorized_sqlstore.get_authenticated_user") as mock_get_user:
+    with patch("llama_stack.core.storage.sqlstore.authorized_sqlstore.get_authenticated_user") as mock_get_user:
         mock_get_user.return_value = user_b
         listed = await s3_provider.openai_list_files()
         assert any(f.id == uploaded.id for f in listed.data)
@@ -80,10 +79,10 @@ async def test_shared_role_allows_access(s3_provider, sample_text_file, op):
     user_x = User("user-x", {"roles": ["shared-role"]})
     user_y = User("user-y", {"roles": ["shared-role"]})
 
-    with patch("llama_stack.providers.utils.sqlstore.authorized_sqlstore.get_authenticated_user") as mock_get_user:
+    with patch("llama_stack.core.storage.sqlstore.authorized_sqlstore.get_authenticated_user") as mock_get_user:
         mock_get_user.return_value = user_x
         uploaded = await s3_provider.openai_upload_file(file=sample_text_file, purpose=OpenAIFilePurpose.ASSISTANTS)
 
-    with patch("llama_stack.providers.utils.sqlstore.authorized_sqlstore.get_authenticated_user") as mock_get_user:
+    with patch("llama_stack.core.storage.sqlstore.authorized_sqlstore.get_authenticated_user") as mock_get_user:
         mock_get_user.return_value = user_y
         await op(s3_provider, uploaded.id)

tests/unit/providers/inference/test_bedrock_adapter.py

@@ -10,9 +10,9 @@ from unittest.mock import AsyncMock, MagicMock
 import pytest
 from openai import AuthenticationError
 
-from llama_stack.apis.inference import OpenAIChatCompletionRequestWithExtraBody
 from llama_stack.providers.remote.inference.bedrock.bedrock import BedrockInferenceAdapter
 from llama_stack.providers.remote.inference.bedrock.config import BedrockConfig
+from llama_stack_api import OpenAIChatCompletionRequestWithExtraBody
 
 
 def test_adapter_initialization():

tests/unit/providers/inference/test_remote_vllm.py

@@ -10,7 +10,13 @@ from unittest.mock import AsyncMock, MagicMock, PropertyMock, patch
 
 import pytest
 
-from llama_stack.apis.inference import (
+from llama_stack.core.routers.inference import InferenceRouter
+from llama_stack.core.routing_tables.models import ModelsRoutingTable
+from llama_stack.providers.remote.inference.vllm.config import VLLMInferenceAdapterConfig
+from llama_stack.providers.remote.inference.vllm.vllm import VLLMInferenceAdapter
+from llama_stack_api import (
+    HealthStatus,
+    Model,
     OpenAIAssistantMessageParam,
     OpenAIChatCompletion,
     OpenAIChatCompletionRequestWithExtraBody,
@@ -20,12 +26,6 @@ from llama_stack.apis.inference import (
     OpenAICompletionRequestWithExtraBody,
     ToolChoice,
 )
-from llama_stack.apis.models import Model
-from llama_stack.core.routers.inference import InferenceRouter
-from llama_stack.core.routing_tables.models import ModelsRoutingTable
-from llama_stack.providers.datatypes import HealthStatus
-from llama_stack.providers.remote.inference.vllm.config import VLLMInferenceAdapterConfig
-from llama_stack.providers.remote.inference.vllm.vllm import VLLMInferenceAdapter
 
 # These are unit test for the remote vllm provider
 # implementation. This should only contain tests which are specific to

tests/unit/providers/inline/agents/meta_reference/responses/test_streaming.py

@@ -8,11 +8,11 @@ from unittest.mock import AsyncMock
 
 import pytest
 
-from llama_stack.apis.tools import ToolDef
 from llama_stack.providers.inline.agents.meta_reference.responses.streaming import (
     convert_tooldef_to_chat_tool,
 )
 from llama_stack.providers.inline.agents.meta_reference.responses.types import ChatCompletionContext
+from llama_stack_api import ToolDef
 
 
 @pytest.fixture

tests/unit/providers/inline/inference/__init__.py

@@ -0,0 +1,5 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.

tests/unit/providers/inline/inference/test_meta_reference.py

@@ -0,0 +1,44 @@
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+#
+# This source code is licensed under the terms described in the LICENSE file in
+# the root directory of this source tree.
+
+from unittest.mock import Mock
+
+import pytest
+
+from llama_stack.providers.inline.inference.meta_reference.model_parallel import (
+    ModelRunner,
+)
+
+
+class TestModelRunner:
+    """Test ModelRunner task dispatching for model-parallel inference."""
+
+    def test_chat_completion_task_dispatch(self):
+        """Verify ModelRunner correctly dispatches chat_completion tasks."""
+        # Create a mock generator
+        mock_generator = Mock()
+        mock_generator.chat_completion = Mock(return_value=iter([]))
+
+        runner = ModelRunner(mock_generator)
+
+        # Create a chat_completion task
+        fake_params = {"model": "test"}
+        fake_messages = [{"role": "user", "content": "test"}]
+        task = ("chat_completion", [fake_params, fake_messages])
+
+        # Execute task
+        runner(task)
+
+        # Verify chat_completion was called with correct arguments
+        mock_generator.chat_completion.assert_called_once_with(fake_params, fake_messages)
+
+    def test_invalid_task_type_raises_error(self):
+        """Verify ModelRunner rejects invalid task types."""
+        mock_generator = Mock()
+        runner = ModelRunner(mock_generator)
+
+        with pytest.raises(ValueError, match="Unexpected task type"):
+            runner(("invalid_task", []))

tests/unit/providers/nvidia/test_datastore.py

@@ -9,10 +9,9 @@ from unittest.mock import patch
 
 import pytest
 
-from llama_stack.apis.datasets import Dataset, DatasetPurpose, URIDataSource
-from llama_stack.apis.resource import ResourceType
 from llama_stack.providers.remote.datasetio.nvidia.config import NvidiaDatasetIOConfig
 from llama_stack.providers.remote.datasetio.nvidia.datasetio import NvidiaDatasetIOAdapter
+from llama_stack_api import Dataset, DatasetPurpose, ResourceType, URIDataSource
 
 
 @pytest.fixture

tests/unit/providers/nvidia/test_eval.py

@@ -9,14 +9,20 @@ from unittest.mock import MagicMock, patch
 
 import pytest
 
-from llama_stack.apis.benchmarks import Benchmark
-from llama_stack.apis.common.job_types import Job, JobStatus
-from llama_stack.apis.eval.eval import BenchmarkConfig, EvaluateResponse, ModelCandidate, SamplingParams
-from llama_stack.apis.inference.inference import TopPSamplingStrategy
-from llama_stack.apis.resource import ResourceType
 from llama_stack.models.llama.sku_types import CoreModelId
 from llama_stack.providers.remote.eval.nvidia.config import NVIDIAEvalConfig
 from llama_stack.providers.remote.eval.nvidia.eval import NVIDIAEvalImpl
+from llama_stack_api import (
+    Benchmark,
+    BenchmarkConfig,
+    EvaluateResponse,
+    Job,
+    JobStatus,
+    ModelCandidate,
+    ResourceType,
+    SamplingParams,
+    TopPSamplingStrategy,
+)
 
 MOCK_DATASET_ID = "default/test-dataset"
 MOCK_BENCHMARK_ID = "test-benchmark"

tests/unit/providers/nvidia/test_parameters.py

@@ -10,7 +10,12 @@ from unittest.mock import patch
 
 import pytest
 
-from llama_stack.apis.post_training.post_training import (
+from llama_stack.core.library_client import convert_pydantic_to_json_value
+from llama_stack.providers.remote.post_training.nvidia.post_training import (
+    NvidiaPostTrainingAdapter,
+    NvidiaPostTrainingConfig,
+)
+from llama_stack_api import (
     DataConfig,
     DatasetFormat,
     EfficiencyConfig,
@@ -19,11 +24,6 @@ from llama_stack.apis.post_training.post_training import (
     OptimizerType,
     TrainingConfig,
 )
-from llama_stack.core.library_client import convert_pydantic_to_json_value
-from llama_stack.providers.remote.post_training.nvidia.post_training import (
-    NvidiaPostTrainingAdapter,
-    NvidiaPostTrainingConfig,
-)
 
 
 class TestNvidiaParameters:

tests/unit/providers/nvidia/test_rerank_inference.py

@@ -9,10 +9,10 @@ from unittest.mock import AsyncMock, MagicMock, patch
 import aiohttp
 import pytest
 
-from llama_stack.apis.models import ModelType
 from llama_stack.providers.remote.inference.nvidia.config import NVIDIAConfig
 from llama_stack.providers.remote.inference.nvidia.nvidia import NVIDIAInferenceAdapter
 from llama_stack.providers.utils.inference.openai_mixin import OpenAIMixin
+from llama_stack_api import ModelType
 
 
 class MockResponse:

tests/unit/providers/nvidia/test_safety.py

@@ -10,13 +10,16 @@ from unittest.mock import AsyncMock, MagicMock, patch
 
 import pytest
 
-from llama_stack.apis.inference import CompletionMessage, UserMessage
-from llama_stack.apis.resource import ResourceType
-from llama_stack.apis.safety import RunShieldResponse, ViolationLevel
-from llama_stack.apis.shields import Shield
-from llama_stack.models.llama.datatypes import StopReason
 from llama_stack.providers.remote.safety.nvidia.config import NVIDIASafetyConfig
 from llama_stack.providers.remote.safety.nvidia.nvidia import NVIDIASafetyAdapter
+from llama_stack_api import (
+    OpenAIAssistantMessageParam,
+    OpenAIUserMessageParam,
+    ResourceType,
+    RunShieldResponse,
+    Shield,
+    ViolationLevel,
+)
 
 
 class FakeNVIDIASafetyAdapter(NVIDIASafetyAdapter):
@@ -136,11 +139,9 @@ async def test_run_shield_allowed(nvidia_adapter, mock_guardrails_post):
 
     # Run the shield
     messages = [
-        UserMessage(role="user", content="Hello, how are you?"),
-        CompletionMessage(
-            role="assistant",
+        OpenAIUserMessageParam(content="Hello, how are you?"),
+        OpenAIAssistantMessageParam(
             content="I'm doing well, thank you for asking!",
-            stop_reason=StopReason.end_of_message,
             tool_calls=[],
         ),
     ]
@@ -191,13 +192,10 @@ async def test_run_shield_blocked(nvidia_adapter, mock_guardrails_post):
     # Mock Guardrails API response
     mock_guardrails_post.return_value = {"status": "blocked", "rails_status": {"reason": "harmful_content"}}
 
-    # Run the shield
     messages = [
-        UserMessage(role="user", content="Hello, how are you?"),
-        CompletionMessage(
-            role="assistant",
+        OpenAIUserMessageParam(content="Hello, how are you?"),
+        OpenAIAssistantMessageParam(
             content="I'm doing well, thank you for asking!",
-            stop_reason=StopReason.end_of_message,
             tool_calls=[],
         ),
     ]
@@ -243,7 +241,7 @@ async def test_run_shield_not_found(nvidia_adapter, mock_guardrails_post):
     adapter.shield_store.get_shield.return_value = None
 
     messages = [
-        UserMessage(role="user", content="Hello, how are you?"),
+        OpenAIUserMessageParam(content="Hello, how are you?"),
     ]
 
     with pytest.raises(ValueError):
@@ -274,11 +272,9 @@ async def test_run_shield_http_error(nvidia_adapter, mock_guardrails_post):
 
     # Running the shield should raise an exception
     messages = [
-        UserMessage(role="user", content="Hello, how are you?"),
-        CompletionMessage(
-            role="assistant",
+        OpenAIUserMessageParam(content="Hello, how are you?"),
+        OpenAIAssistantMessageParam(
             content="I'm doing well, thank you for asking!",
-            stop_reason=StopReason.end_of_message,
             tool_calls=[],
         ),
     ]

tests/unit/providers/nvidia/test_supervised_fine_tuning.py

@@ -10,15 +10,6 @@ from unittest.mock import patch
 
 import pytest
 
-from llama_stack.apis.post_training.post_training import (
-    DataConfig,
-    DatasetFormat,
-    LoraFinetuningConfig,
-    OptimizerConfig,
-    OptimizerType,
-    QATFinetuningConfig,
-    TrainingConfig,
-)
 from llama_stack.core.library_client import convert_pydantic_to_json_value
 from llama_stack.providers.remote.post_training.nvidia.post_training import (
     ListNvidiaPostTrainingJobs,
@@ -27,6 +18,15 @@ from llama_stack.providers.remote.post_training.nvidia.post_training import (
     NvidiaPostTrainingJob,
     NvidiaPostTrainingJobStatusResponse,
 )
+from llama_stack_api import (
+    DataConfig,
+    DatasetFormat,
+    LoraFinetuningConfig,
+    OptimizerConfig,
+    OptimizerType,
+    QATFinetuningConfig,
+    TrainingConfig,
+)
 
 
 @pytest.fixture

tests/unit/providers/test_bedrock.py

@@ -7,9 +7,9 @@
 from types import SimpleNamespace
 from unittest.mock import AsyncMock, PropertyMock, patch
 
-from llama_stack.apis.inference import OpenAIChatCompletionRequestWithExtraBody
 from llama_stack.providers.remote.inference.bedrock.bedrock import BedrockInferenceAdapter
 from llama_stack.providers.remote.inference.bedrock.config import BedrockConfig
+from llama_stack_api import OpenAIChatCompletionRequestWithExtraBody
 
 
 def test_can_create_adapter():