Merge branch 'main' into responses-and-safety

.github/workflows/integration-auth-tests.yml

@@ -92,7 +92,8 @@ jobs:
         run: |
           echo "Waiting for Llama Stack server..."
           for i in {1..30}; do
-            if curl -s -L -H "Authorization: Bearer $(cat llama-stack-auth-token)" http://localhost:8321/v1/health | grep -q "OK"; then
+            # Note: /v1/health does not require authentication
+            if curl -s -L http://localhost:8321/v1/health | grep -q "OK"; then
               echo "Llama Stack server is up!"
               if grep -q "Enabling authentication with provider: ${{ matrix.auth-provider }}" server.log; then
                 echo "Llama Stack server is configured to use ${{ matrix.auth-provider }} auth"
@@ -111,4 +112,27 @@ jobs:
 
       - name: Test auth
         run: |
-          curl -s -L -H "Authorization: Bearer $(cat llama-stack-auth-token)" http://127.0.0.1:8321/v1/providers|jq
+          echo "Testing /v1/version without token (should succeed)..."
+          if curl -s -L -o /dev/null -w "%{http_code}" http://127.0.0.1:8321/v1/version | grep -q "200"; then
+            echo "/v1/version accessible without token (200)"
+          else
+            echo "/v1/version returned non-200 status without token"
+            exit 1
+          fi
+
+          echo "Testing /v1/providers without token (should fail with 401)..."
+          if curl -s -L -o /dev/null -w "%{http_code}" http://127.0.0.1:8321/v1/providers | grep -q "401"; then
+            echo "/v1/providers blocked without token (401)"
+          else
+            echo "/v1/providers did not return 401 without token"
+            exit 1
+          fi
+
+          echo "Testing /v1/providers with valid token (should succeed)..."
+          curl -s -L -H "Authorization: Bearer $(cat llama-stack-auth-token)" http://127.0.0.1:8321/v1/providers | jq
+          if [ $? -eq 0 ]; then
+            echo "/v1/providers accessible with valid token"
+          else
+            echo "/v1/providers failed with valid token"
+            exit 1
+          fi

llama_stack/apis/inspect/inspect.py

@@ -73,7 +73,7 @@ class Inspect(Protocol):
         """
         ...
 
-    @webmethod(route="/health", method="GET", level=LLAMA_STACK_API_V1)
+    @webmethod(route="/health", method="GET", level=LLAMA_STACK_API_V1, require_authentication=False)
     async def health(self) -> HealthInfo:
         """Get health status.
 
@@ -83,7 +83,7 @@ class Inspect(Protocol):
         """
         ...
 
-    @webmethod(route="/version", method="GET", level=LLAMA_STACK_API_V1)
+    @webmethod(route="/version", method="GET", level=LLAMA_STACK_API_V1, require_authentication=False)
     async def version(self) -> VersionInfo:
         """Get version.
 

llama_stack/core/server/auth.py

@@ -27,6 +27,11 @@ class AuthenticationMiddleware:
     3. Extracts user attributes from the provider's response
     4. Makes these attributes available to the route handlers for access control
 
+    Unauthenticated Access:
+    Endpoints can opt out of authentication by setting require_authentication=False
+    in their @webmethod decorator. This is typically used for operational endpoints
+    like /health and /version to support monitoring, load balancers, and observability tools.
+
     The middleware supports multiple authentication providers through the AuthProvider interface:
     - Kubernetes: Validates tokens against the Kubernetes API server
     - Custom: Validates tokens against a custom endpoint
@@ -88,7 +93,26 @@ class AuthenticationMiddleware:
 
     async def __call__(self, scope, receive, send):
         if scope["type"] == "http":
-            # First, handle authentication
+            # Find the route and check if authentication is required
+            path = scope.get("path", "")
+            method = scope.get("method", hdrs.METH_GET)
+
+            if not hasattr(self, "route_impls"):
+                self.route_impls = initialize_route_impls(self.impls)
+
+            webmethod = None
+            try:
+                _, _, _, webmethod = find_matching_route(method, path, self.route_impls)
+            except ValueError:
+                # If no matching endpoint is found, pass here to run auth anyways
+                pass
+
+            # If webmethod explicitly sets require_authentication=False, allow without auth
+            if webmethod and webmethod.require_authentication is False:
+                logger.debug(f"Allowing unauthenticated access to endpoint: {path}")
+                return await self.app(scope, receive, send)
+
+            # Handle authentication
             headers = dict(scope.get("headers", []))
             auth_header = headers.get(b"authorization", b"").decode()
 
@@ -127,19 +151,7 @@ class AuthenticationMiddleware:
             )
 
             # Scope-based API access control
-            path = scope.get("path", "")
-            method = scope.get("method", hdrs.METH_GET)
-
-            if not hasattr(self, "route_impls"):
-                self.route_impls = initialize_route_impls(self.impls)
-
-            try:
-                _, _, _, webmethod = find_matching_route(method, path, self.route_impls)
-            except ValueError:
-                # If no matching endpoint is found, pass through to FastAPI
-                return await self.app(scope, receive, send)
-
-            if webmethod.required_scope:
+            if webmethod and webmethod.required_scope:
                 user = user_from_scope(scope)
                 if not _has_required_scope(webmethod.required_scope, user):
                     return await self._send_auth_error(

llama_stack/providers/inline/agents/meta_reference/responses/streaming.py

@@ -41,12 +41,16 @@ from llama_stack.apis.agents.openai_responses import (
     OpenAIResponseOutputMessageFunctionToolCall,
     OpenAIResponseOutputMessageMCPListTools,
     OpenAIResponseText,
+    OpenAIResponseUsage,
+    OpenAIResponseUsageInputTokensDetails,
+    OpenAIResponseUsageOutputTokensDetails,
     WebSearchToolTypes,
 )
 from llama_stack.apis.inference import (
     Inference,
     OpenAIAssistantMessageParam,
     OpenAIChatCompletion,
+    OpenAIChatCompletionChunk,
     OpenAIChatCompletionToolCall,
     OpenAIChoice,
     OpenAIMessageParam,
@@ -116,43 +120,8 @@ class StreamingResponseOrchestrator:
         self.final_messages: list[OpenAIMessageParam] = []
         # mapping for annotations
         self.citation_files: dict[str, str] = {}
-        # Track accumulated text for shield validation
-        self.accumulated_text = ""
-        # Track if we've sent a refusal response
-        self.violation_detected = False
-
-    async def _check_output_stream_safety(self, text_delta: str) -> str | None:
-        """Check streaming text content against shields. Returns violation message if blocked."""
-        if not self.shield_ids:
-            return None
-
-        self.accumulated_text += text_delta
-
-        # Check accumulated text periodically for violations (every 50 characters or at word boundaries)
-        if len(self.accumulated_text) > 50 or text_delta.endswith((" ", "\n", ".", "!", "?")):
-            temp_messages = [{"role": "assistant", "content": self.accumulated_text}]
-            messages = convert_openai_to_inference_messages(temp_messages)
-
-            try:
-                await run_multiple_shields(self.safety_api, messages, self.shield_ids)
-            except SafetyException as e:
-                logger.info(f"Output shield violation: {e.violation.user_message}")
-                return e.violation.user_message or "Generated content blocked by safety shields"
-
-    async def _create_refusal_response(self, violation_message: str) -> OpenAIResponseObjectStream:
-        """Create a refusal response to replace streaming content."""
-        refusal_content = OpenAIResponseContentPartRefusal(refusal=violation_message)
-
-        # Create a completed refusal response
-        refusal_response = OpenAIResponseObject(
-            id=self.response_id,
-            created_at=self.created_at,
-            model=self.ctx.model,
-            status="completed",
-            output=[OpenAIResponseMessage(role="assistant", content=[refusal_content], type="message")],
-        )
-
-        return OpenAIResponseObjectStreamResponseCompleted(response=refusal_response)
+        # Track accumulated usage across all inference calls
+        self.accumulated_usage: OpenAIResponseUsage | None = None
 
     def _clone_outputs(self, outputs: list[OpenAIResponseOutput]) -> list[OpenAIResponseOutput]:
         cloned: list[OpenAIResponseOutput] = []
@@ -180,6 +149,7 @@ class StreamingResponseOrchestrator:
             text=self.text,
             tools=self.ctx.available_tools(),
             error=error,
+            usage=self.accumulated_usage,
         )
 
     async def create_response(self) -> AsyncIterator[OpenAIResponseObjectStream]:
@@ -217,6 +187,9 @@ class StreamingResponseOrchestrator:
                     stream=True,
                     temperature=self.ctx.temperature,
                     response_format=response_format,
+                    stream_options={
+                        "include_usage": True,
+                    },
                 )
 
                 # Process streaming chunks and build complete response
@@ -352,6 +325,51 @@ class StreamingResponseOrchestrator:
 
         return function_tool_calls, non_function_tool_calls, approvals, next_turn_messages
 
+    def _accumulate_chunk_usage(self, chunk: OpenAIChatCompletionChunk) -> None:
+        """Accumulate usage from a streaming chunk into the response usage format."""
+        if not chunk.usage:
+            return
+
+        if self.accumulated_usage is None:
+            # Convert from chat completion format to response format
+            self.accumulated_usage = OpenAIResponseUsage(
+                input_tokens=chunk.usage.prompt_tokens,
+                output_tokens=chunk.usage.completion_tokens,
+                total_tokens=chunk.usage.total_tokens,
+                input_tokens_details=(
+                    OpenAIResponseUsageInputTokensDetails(cached_tokens=chunk.usage.prompt_tokens_details.cached_tokens)
+                    if chunk.usage.prompt_tokens_details
+                    else None
+                ),
+                output_tokens_details=(
+                    OpenAIResponseUsageOutputTokensDetails(
+                        reasoning_tokens=chunk.usage.completion_tokens_details.reasoning_tokens
+                    )
+                    if chunk.usage.completion_tokens_details
+                    else None
+                ),
+            )
+        else:
+            # Accumulate across multiple inference calls
+            self.accumulated_usage = OpenAIResponseUsage(
+                input_tokens=self.accumulated_usage.input_tokens + chunk.usage.prompt_tokens,
+                output_tokens=self.accumulated_usage.output_tokens + chunk.usage.completion_tokens,
+                total_tokens=self.accumulated_usage.total_tokens + chunk.usage.total_tokens,
+                # Use latest non-null details
+                input_tokens_details=(
+                    OpenAIResponseUsageInputTokensDetails(cached_tokens=chunk.usage.prompt_tokens_details.cached_tokens)
+                    if chunk.usage.prompt_tokens_details
+                    else self.accumulated_usage.input_tokens_details
+                ),
+                output_tokens_details=(
+                    OpenAIResponseUsageOutputTokensDetails(
+                        reasoning_tokens=chunk.usage.completion_tokens_details.reasoning_tokens
+                    )
+                    if chunk.usage.completion_tokens_details
+                    else self.accumulated_usage.output_tokens_details
+                ),
+            )
+
     async def _process_streaming_chunks(
         self, completion_result, output_messages: list[OpenAIResponseOutput]
     ) -> AsyncIterator[OpenAIResponseObjectStream | ChatCompletionResult]:
@@ -377,6 +395,10 @@ class StreamingResponseOrchestrator:
             chat_response_id = chunk.id
             chunk_created = chunk.created
             chunk_model = chunk.model
+
+            # Accumulate usage from chunks (typically in final chunk with stream_options)
+            self._accumulate_chunk_usage(chunk)
+
             for chunk_choice in chunk.choices:
                 # Emit incremental text content as delta events
                 if chunk_choice.delta.content:

llama_stack/providers/inline/batches/reference/batches.py

@@ -178,9 +178,9 @@ class ReferenceBatchesImpl(Batches):
 
         # TODO: set expiration time for garbage collection
 
-        if endpoint not in ["/v1/chat/completions", "/v1/completions"]:
+        if endpoint not in ["/v1/chat/completions", "/v1/completions", "/v1/embeddings"]:
             raise ValueError(
-                f"Invalid endpoint: {endpoint}. Supported values: /v1/chat/completions, /v1/completions. Code: invalid_value. Param: endpoint",
+                f"Invalid endpoint: {endpoint}. Supported values: /v1/chat/completions, /v1/completions, /v1/embeddings. Code: invalid_value. Param: endpoint",
             )
 
         if completion_window != "24h":
@@ -425,18 +425,23 @@ class ReferenceBatchesImpl(Batches):
                             valid = False
 
                         if batch.endpoint == "/v1/chat/completions":
-                            required_params = [
+                            required_params: list[tuple[str, Any, str]] = [
                                 ("model", str, "a string"),
                                 # messages is specific to /v1/chat/completions
                                 # we could skip validating messages here and let inference fail. however,
                                 # that would be a very expensive way to find out messages is wrong.
                                 ("messages", list, "an array"),  # TODO: allow messages to be a string?
                             ]
-                        else:  # /v1/completions
+                        elif batch.endpoint == "/v1/completions":
                             required_params = [
                                 ("model", str, "a string"),
                                 ("prompt", str, "a string"),  # TODO: allow prompt to be a list of strings??
                             ]
+                        else:  # /v1/embeddings
+                            required_params = [
+                                ("model", str, "a string"),
+                                ("input", (str, list), "a string or array of strings"),
+                            ]
 
                         for param, expected_type, type_string in required_params:
                             if param not in body:
@@ -614,7 +619,7 @@ class ReferenceBatchesImpl(Batches):
                         "body": chat_response.model_dump_json(),
                     },
                 }
-            else:  # /v1/completions
+            elif request.url == "/v1/completions":
                 completion_response = await self.inference_api.openai_completion(**request.body)
 
                 # this is for mypy, we don't allow streaming so we'll get the right type
@@ -630,6 +635,20 @@ class ReferenceBatchesImpl(Batches):
                         "body": completion_response.model_dump_json(),
                     },
                 }
+            else:  # /v1/embeddings
+                embeddings_response = await self.inference_api.openai_embeddings(**request.body)
+                assert hasattr(embeddings_response, "model_dump_json"), (
+                    "Embeddings response must have model_dump_json method"
+                )
+                return {
+                    "id": request_id,
+                    "custom_id": request.custom_id,
+                    "response": {
+                        "status_code": 200,
+                        "request_id": request_id,  # TODO: should this be different?
+                        "body": embeddings_response.model_dump_json(),
+                    },
+                }
         except Exception as e:
             logger.info(f"Error processing request {request.custom_id} in batch {batch_id}: {e}")
             return {

llama_stack/schema_utils.py

@@ -61,6 +61,7 @@ class WebMethod:
     descriptive_name: str | None = None
     required_scope: str | None = None
     deprecated: bool | None = False
+    require_authentication: bool | None = True
 
 
 CallableT = TypeVar("CallableT", bound=Callable[..., Any])
@@ -77,6 +78,7 @@ def webmethod(
     descriptive_name: str | None = None,
     required_scope: str | None = None,
     deprecated: bool | None = False,
+    require_authentication: bool | None = True,
 ) -> Callable[[CallableT], CallableT]:
     """
     Decorator that supplies additional metadata to an endpoint operation function.
@@ -86,6 +88,7 @@ def webmethod(
     :param request_examples: Sample requests that the operation might take. Pass a list of objects, not JSON.
     :param response_examples: Sample responses that the operation might produce. Pass a list of objects, not JSON.
     :param required_scope: Required scope for this endpoint (e.g., 'monitoring.viewer').
+    :param require_authentication: Whether this endpoint requires authentication (default True).
     """
 
     def wrap(func: CallableT) -> CallableT:
@@ -100,6 +103,7 @@ def webmethod(
             descriptive_name=descriptive_name,
             required_scope=required_scope,
             deprecated=deprecated,
+            require_authentication=require_authentication if require_authentication is not None else True,
         )
 
         # Store all webmethods in a list to support multiple decorators

tests/integration/agents/recordings/2172059863d4d17e7525483102a6753617b60a8c33ece637db18061d23086536.json

@@ -0,0 +1,130 @@
+{
+  "test_id": "tests/integration/agents/test_openai_responses.py::test_responses_store[openai_client-txt=ollama/llama3.2:3b-instruct-fp16-tools1-True]",
+  "request": {
+    "method": "POST",
+    "url": "http://0.0.0.0:11434/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "llama3.2:3b-instruct-fp16",
+      "messages": [
+        {
+          "role": "user",
+          "content": "What's the weather in Tokyo? YOU MUST USE THE get_weather function to get the weather."
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      },
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "type": "function",
+            "name": "get_weather",
+            "description": "Get the weather in a given city",
+            "parameters": {
+              "type": "object",
+              "properties": {
+                "city": {
+                  "type": "string",
+                  "description": "The city to get the weather for"
+                }
+              }
+            },
+            "strict": null
+          }
+        }
+      ]
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "llama3.2:3b-instruct-fp16"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-2172059863d4",
+          "choices": [
+            {
+              "delta": {
+                "content": "",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": "call_os3xa9go",
+                    "function": {
+                      "arguments": "{\"city\":\"Tokyo\"}",
+                      "name": "get_weather"
+                    },
+                    "type": "function"
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": null
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-2172059863d4",
+          "choices": [
+            {
+              "delta": {
+                "content": "",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": "tool_calls",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": null
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-2172059863d4",
+          "choices": [],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": {
+            "completion_tokens": 15,
+            "prompt_tokens": 179,
+            "total_tokens": 194,
+            "completion_tokens_details": null,
+            "prompt_tokens_details": null
+          }
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/agents/recordings/4da32cdf48ae4c3b381e3557edb99afa0ff16a609aaa941737e99606961a6a07.json

@@ -0,0 +1,272 @@
+{
+  "test_id": "tests/integration/agents/test_openai_responses.py::test_list_response_input_items_with_limit_and_order[txt=ollama/llama3.2:3b-instruct-fp16]",
+  "request": {
+    "method": "POST",
+    "url": "http://0.0.0.0:11434/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "llama3.2:3b-instruct-fp16",
+      "messages": [
+        {
+          "role": "user",
+          "content": "Message A: What is the capital of France?"
+        },
+        {
+          "role": "assistant",
+          "content": "The capital of France is Paris."
+        },
+        {
+          "role": "user",
+          "content": "Message B: What about Spain?"
+        },
+        {
+          "role": "assistant",
+          "content": "The capital of Spain is Madrid."
+        },
+        {
+          "role": "user",
+          "content": "Message C: And Italy?"
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      }
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "llama3.2:3b-instruct-fp16"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-4da32cdf48ae",
+          "choices": [
+            {
+              "delta": {
+                "content": "The",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": null
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-4da32cdf48ae",
+          "choices": [
+            {
+              "delta": {
+                "content": " capital",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": null
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-4da32cdf48ae",
+          "choices": [
+            {
+              "delta": {
+                "content": " of",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": null
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-4da32cdf48ae",
+          "choices": [
+            {
+              "delta": {
+                "content": " Italy",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": null
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-4da32cdf48ae",
+          "choices": [
+            {
+              "delta": {
+                "content": " is",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": null
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-4da32cdf48ae",
+          "choices": [
+            {
+              "delta": {
+                "content": " Rome",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": null
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-4da32cdf48ae",
+          "choices": [
+            {
+              "delta": {
+                "content": ".",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": null
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-4da32cdf48ae",
+          "choices": [
+            {
+              "delta": {
+                "content": "",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": "stop",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": null
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-4da32cdf48ae",
+          "choices": [],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": {
+            "completion_tokens": 8,
+            "prompt_tokens": 82,
+            "total_tokens": 90,
+            "completion_tokens_details": null,
+            "prompt_tokens_details": null
+          }
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/agents/recordings/585a2cf2c22b0db155a6a94052836b34c42f68bf04b1b2cb74ddf00943c0442d.json

@@ -0,0 +1,122 @@
+{
+  "test_id": "tests/integration/agents/test_openai_responses.py::test_function_call_output_response_with_none_arguments[txt=ollama/llama3.2:3b-instruct-fp16]",
+  "request": {
+    "method": "POST",
+    "url": "http://0.0.0.0:11434/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "llama3.2:3b-instruct-fp16",
+      "messages": [
+        {
+          "role": "user",
+          "content": "what's the current time? You MUST call the `get_current_time` function to find out."
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      },
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "type": "function",
+            "name": "get_current_time",
+            "description": "Get the current time",
+            "parameters": {},
+            "strict": null
+          }
+        }
+      ]
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "llama3.2:3b-instruct-fp16"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-585a2cf2c22b",
+          "choices": [
+            {
+              "delta": {
+                "content": "",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": "call_lqrdy0rt",
+                    "function": {
+                      "arguments": "{}",
+                      "name": "get_current_time"
+                    },
+                    "type": "function"
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": null
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-585a2cf2c22b",
+          "choices": [
+            {
+              "delta": {
+                "content": "",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": "tool_calls",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": null
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-585a2cf2c22b",
+          "choices": [],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": {
+            "completion_tokens": 14,
+            "prompt_tokens": 161,
+            "total_tokens": 175,
+            "completion_tokens_details": null,
+            "prompt_tokens_details": null
+          }
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/agents/recordings/5e4ea6b89ff437f9680dd9e31aaa2a5e46d3cc8849c5cc823142d4c0bb2c4c40.json

@@ -0,0 +1,256 @@
+{
+  "test_id": "tests/integration/agents/test_openai_responses.py::test_list_response_input_items[openai_client-txt=ollama/llama3.2:3b-instruct-fp16]",
+  "request": {
+    "method": "POST",
+    "url": "http://0.0.0.0:11434/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "llama3.2:3b-instruct-fp16",
+      "messages": [
+        {
+          "role": "user",
+          "content": "What is the capital of France?"
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      }
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "llama3.2:3b-instruct-fp16"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5e4ea6b89ff4",
+          "choices": [
+            {
+              "delta": {
+                "content": "The",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": null
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5e4ea6b89ff4",
+          "choices": [
+            {
+              "delta": {
+                "content": " capital",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": null
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5e4ea6b89ff4",
+          "choices": [
+            {
+              "delta": {
+                "content": " of",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": null
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5e4ea6b89ff4",
+          "choices": [
+            {
+              "delta": {
+                "content": " France",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": null
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5e4ea6b89ff4",
+          "choices": [
+            {
+              "delta": {
+                "content": " is",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": null
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5e4ea6b89ff4",
+          "choices": [
+            {
+              "delta": {
+                "content": " Paris",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": null
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5e4ea6b89ff4",
+          "choices": [
+            {
+              "delta": {
+                "content": ".",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": null
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5e4ea6b89ff4",
+          "choices": [
+            {
+              "delta": {
+                "content": "",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": "stop",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": null
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5e4ea6b89ff4",
+          "choices": [],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": {
+            "completion_tokens": 8,
+            "prompt_tokens": 32,
+            "total_tokens": 40,
+            "completion_tokens_details": null,
+            "prompt_tokens_details": null
+          }
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/agents/recordings/5edf2f0b7a9c875e80e4719f71a1daa94c1287acf164cd81ddd51843d05be718.json

@@ -0,0 +1,130 @@
+{
+  "test_id": "tests/integration/agents/test_openai_responses.py::test_responses_store[openai_client-txt=ollama/llama3.2:3b-instruct-fp16-tools1-False]",
+  "request": {
+    "method": "POST",
+    "url": "http://0.0.0.0:11434/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "llama3.2:3b-instruct-fp16",
+      "messages": [
+        {
+          "role": "user",
+          "content": "What's the weather in Tokyo? YOU MUST USE THE get_weather function to get the weather."
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      },
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "type": "function",
+            "name": "get_weather",
+            "description": "Get the weather in a given city",
+            "parameters": {
+              "type": "object",
+              "properties": {
+                "city": {
+                  "type": "string",
+                  "description": "The city to get the weather for"
+                }
+              }
+            },
+            "strict": null
+          }
+        }
+      ]
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "llama3.2:3b-instruct-fp16"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5edf2f0b7a9c",
+          "choices": [
+            {
+              "delta": {
+                "content": "",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": [
+                  {
+                    "index": 0,
+                    "id": "call_4ibtjudr",
+                    "function": {
+                      "arguments": "{\"city\":\"Tokyo\"}",
+                      "name": "get_weather"
+                    },
+                    "type": "function"
+                  }
+                ]
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": null
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5edf2f0b7a9c",
+          "choices": [
+            {
+              "delta": {
+                "content": "",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": "tool_calls",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": null
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5edf2f0b7a9c",
+          "choices": [],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": {
+            "completion_tokens": 15,
+            "prompt_tokens": 179,
+            "total_tokens": 194,
+            "completion_tokens_details": null,
+            "prompt_tokens_details": null
+          }
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/agents/recordings/f4894c38303b5bf152c329b51c293cbe58f89b8ef772670856c898c8b56090cf.json

@@ -0,0 +1,256 @@
+{
+  "test_id": "tests/integration/agents/test_openai_responses.py::test_list_response_input_items[client_with_models-txt=ollama/llama3.2:3b-instruct-fp16]",
+  "request": {
+    "method": "POST",
+    "url": "http://0.0.0.0:11434/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "llama3.2:3b-instruct-fp16",
+      "messages": [
+        {
+          "role": "user",
+          "content": "What is the capital of France?"
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      }
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "llama3.2:3b-instruct-fp16"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-f4894c38303b",
+          "choices": [
+            {
+              "delta": {
+                "content": "The",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": null
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-f4894c38303b",
+          "choices": [
+            {
+              "delta": {
+                "content": " capital",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": null
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-f4894c38303b",
+          "choices": [
+            {
+              "delta": {
+                "content": " of",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": null
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-f4894c38303b",
+          "choices": [
+            {
+              "delta": {
+                "content": " France",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": null
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-f4894c38303b",
+          "choices": [
+            {
+              "delta": {
+                "content": " is",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": null
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-f4894c38303b",
+          "choices": [
+            {
+              "delta": {
+                "content": " Paris",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": null
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-f4894c38303b",
+          "choices": [
+            {
+              "delta": {
+                "content": ".",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": null,
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": null
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-f4894c38303b",
+          "choices": [
+            {
+              "delta": {
+                "content": "",
+                "function_call": null,
+                "refusal": null,
+                "role": "assistant",
+                "tool_calls": null
+              },
+              "finish_reason": "stop",
+              "index": 0,
+              "logprobs": null
+            }
+          ],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": null
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-f4894c38303b",
+          "choices": [],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": {
+            "completion_tokens": 8,
+            "prompt_tokens": 32,
+            "total_tokens": 40,
+            "completion_tokens_details": null,
+            "prompt_tokens_details": null
+          }
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/batches/recordings/37b8d062efc39eaa4ac3438f5102044a29fd1e9b844ead2f066a1fb25c85672c.json

@@ -0,0 +1,422 @@
+{
+  "test_id": "tests/integration/batches/test_batches.py::TestBatchesIntegration::test_batch_e2e_embeddings[emb=ollama/all-minilm:l6-v2]",
+  "request": {
+    "method": "POST",
+    "url": "http://0.0.0.0:11434/v1/v1/embeddings",
+    "headers": {},
+    "body": {
+      "model": "all-minilm:l6-v2",
+      "input": "Hello world",
+      "encoding_format": "float"
+    },
+    "endpoint": "/v1/embeddings",
+    "model": "all-minilm:l6-v2"
+  },
+  "response": {
+    "body": {
+      "__type__": "openai.types.create_embedding_response.CreateEmbeddingResponse",
+      "__data__": {
+        "data": [
+          {
+            "embedding": [
+              -0.034477483,
+              0.030899182,
+              0.0066526434,
+              0.026075281,
+              -0.039411988,
+              -0.16037956,
+              0.06692074,
+              -0.006511468,
+              -0.047467157,
+              0.014774274,
+              0.07094562,
+              0.055527706,
+              0.019183245,
+              -0.026297163,
+              -0.010018651,
+              -0.02694715,
+              0.0223884,
+              -0.02220693,
+              -0.14977267,
+              -0.017530814,
+              0.0075938613,
+              0.054253556,
+              0.0032258728,
+              0.031724673,
+              -0.08466085,
+              -0.029342307,
+              0.05155048,
+              0.048105717,
+              -0.0032670307,
+              -0.05822795,
+              0.041971523,
+              0.022229431,
+              0.1281518,
+              -0.022270948,
+              -0.011725874,
+              0.06294936,
+              -0.032847952,
+              -0.09124354,
+              -0.031128692,
+              0.05274829,
+              0.047067728,
+              -0.08414196,
+              -0.029979317,
+              -0.020692566,
+              0.00949804,
+              -0.0035992558,
+              0.0074442336,
+              0.03928378,
+              0.09326073,
+              -0.0037437282,
+              -0.052663893,
+              -0.058101393,
+              -0.006925679,
+              0.0052269334,
+              0.08290669,
+              0.019312402,
+              0.0062818974,
+              -0.010331665,
+              0.008930684,
+              -0.037712026,
+              -0.045175705,
+              0.023950849,
+              -0.006926045,
+              0.013429504,
+              0.100098,
+              -0.0715888,
+              -0.021700105,
+              0.031693522,
+              -0.05161389,
+              -0.08224763,
+              -0.06577986,
+              -0.009853981,
+              0.005808086,
+              0.07364217,
+              -0.034008067,
+              0.024907362,
+              0.014441484,
+              0.02645124,
+              0.009659713,
+              0.030284341,
+              0.052878983,
+              -0.07536944,
+              0.009890014,
+              0.029907802,
+              0.017498897,
+              0.02313779,
+              0.0018918256,
+              0.0013156217,
+              -0.047173936,
+              -0.011251131,
+              -0.11422648,
+              -0.019960148,
+              0.040278148,
+              0.0022633963,
+              -0.07986738,
+              -0.025357265,
+              0.094500035,
+              -0.029062947,
+              -0.14495483,
+              0.2309815,
+              0.027703581,
+              0.03208736,
+              0.031073036,
+              0.042917974,
+              0.064246915,
+              0.032118786,
+              -0.004844535,
+              0.055775862,
+              -0.03756279,
+              -0.021487191,
+              -0.028432492,
+              -0.028887685,
+              0.03842892,
+              -0.017359573,
+              0.052465834,
+              -0.07493626,
+              -0.031175744,
+              0.021936033,
+              -0.039823197,
+              -0.008681939,
+              0.026978256,
+              -0.048551314,
+              0.011414809,
+              0.029628372,
+              -0.020587107,
+              0.013077965,
+              0.028824588,
+              -3.1978743e-33,
+              0.06475607,
+              -0.018065408,
+              0.05190019,
+              0.12193858,
+              0.028755108,
+              0.008794777,
+              -0.07044016,
+              -0.016856866,
+              0.040675826,
+              0.04222898,
+              0.025450956,
+              0.035772353,
+              -0.049134083,
+              0.0021395232,
+              -0.015527445,
+              0.05065655,
+              -0.04814189,
+              0.03586998,
+              -0.004134139,
+              0.10165314,
+              -0.055980552,
+              -0.010677752,
+              0.011231545,
+              0.09068785,
+              0.004311188,
+              0.035094332,
+              -0.009658399,
+              -0.09383056,
+              0.092755266,
+              0.00799794,
+              -0.0077075018,
+              -0.052119244,
+              -0.01259255,
+              0.0032277475,
+              0.005989667,
+              0.0075889886,
+              0.010571857,
+              -0.08629758,
+              -0.06985891,
+              -0.002511263,
+              -0.091053724,
+              0.0468712,
+              0.05203361,
+              0.0072902967,
+              0.010906411,
+              -0.0052922186,
+              0.013883815,
+              0.021929385,
+              0.0341257,
+              0.060227357,
+              0.00018942523,
+              0.0146624865,
+              -0.07000342,
+              0.028425341,
+              -0.027542787,
+              0.01082086,
+              0.03491755,
+              -0.022430921,
+              0.0096813915,
+              0.07725412,
+              0.021618832,
+              0.114911504,
+              -0.06805403,
+              0.023872944,
+              -0.015999107,
+              -0.017794114,
+              0.06442477,
+              0.03206309,
+              0.050293576,
+              -0.005988605,
+              -0.03376946,
+              0.017821673,
+              0.016567992,
+              0.063335925,
+              0.034753703,
+              0.046586752,
+              0.09789875,
+              -0.006560692,
+              0.025039855,
+              -0.07780643,
+              0.016878096,
+              -0.0010056288,
+              0.02257608,
+              -0.0382721,
+              0.09572481,
+              -0.005296001,
+              0.010567662,
+              -0.11538674,
+              -0.013233586,
+              -0.010786205,
+              -0.083147496,
+              0.073254965,
+              0.049377624,
+              -0.009025328,
+              -0.0957893,
+              3.3687185e-33,
+              0.12494067,
+              0.019226579,
+              -0.058172084,
+              -0.035952393,
+              -0.050862074,
+              -0.045700952,
+              -0.0826631,
+              0.14819908,
+              -0.088347495,
+              0.060315337,
+              0.05109269,
+              0.010308115,
+              0.1411753,
+              0.030833788,
+              0.06101746,
+              -0.052806143,
+              0.13661332,
+              0.00917483,
+              -0.017295862,
+              -0.0128495265,
+              -0.007851698,
+              -0.051084496,
+              -0.05235087,
+              0.0076632234,
+              -0.015217299,
+              0.017015414,
+              0.021324545,
+              0.020506723,
+              -0.12004153,
+              0.014523494,
+              0.026743378,
+              0.025221687,
+              -0.04270567,
+              0.00676352,
+              -0.014453511,
+              0.045142446,
+              -0.091383636,
+              -0.019459482,
+              -0.017806036,
+              -0.055010412,
+              -0.05270923,
+              -0.010370778,
+              -0.052053526,
+              0.020918628,
+              -0.080037735,
+              -0.012147244,
+              -0.057777684,
+              0.023249507,
+              -0.007838778,
+              -0.025807643,
+              -0.07987164,
+              -0.020683115,
+              0.04888083,
+              -0.020459235,
+              -0.049192864,
+              0.01407799,
+              -0.063744746,
+              -0.0077936463,
+              0.016429903,
+              -0.025707569,
+              0.013326097,
+              0.026210392,
+              0.009855086,
+              0.06317218,
+              0.0026150644,
+              -0.0065879063,
+              0.0166049,
+              0.032400407,
+              0.038005095,
+              -0.036269873,
+              -0.0069020875,
+              0.00019545198,
+              -0.0017537851,
+              -0.027427403,
+              -0.02801922,
+              0.049696837,
+              -0.028842367,
+              -0.0023814398,
+              0.01481421,
+              0.00976869,
+              0.0057697925,
+              0.01341087,
+              0.00551593,
+              0.037237898,
+              0.007291808,
+              0.040068958,
+              0.08141818,
+              0.07197348,
+              -0.013163506,
+              -0.042782705,
+              -0.010938265,
+              0.0049547236,
+              -0.00923014,
+              0.035068717,
+              -0.051007,
+              -1.5708556e-08,
+              -0.088558294,
+              0.02391312,
+              -0.016132735,
+              0.03169382,
+              0.027184812,
+              0.052484553,
+              -0.047118798,
+              -0.058789898,
+              -0.063239954,
+              0.040775288,
+              0.049807984,
+              0.106462926,
+              -0.07448737,
+              -0.012401869,
+              0.018361589,
+              0.039486438,
+              -0.024830224,
+              0.014500051,
+              -0.03712332,
+              0.020043189,
+              8.399218e-05,
+              0.009852795,
+              0.024823224,
+              -0.05252818,
+              0.02932855,
+              -0.0871494,
+              -0.01447227,
+              0.025996566,
+              -0.018731978,
+              -0.07618361,
+              0.03505914,
+              0.10363578,
+              -0.0280213,
+              0.012769872,
+              -0.076482065,
+              -0.018743375,
+              0.024961015,
+              0.08152011,
+              0.06866303,
+              -0.06411612,
+              -0.08387694,
+              0.061479986,
+              -0.03345598,
+              -0.10615398,
+              -0.040166635,
+              0.032536518,
+              0.076652974,
+              -0.07297006,
+              0.00039833272,
+              -0.0409393,
+              -0.07580284,
+              0.027465926,
+              0.07468789,
+              0.017779494,
+              0.09106629,
+              0.11033428,
+              0.00065298256,
+              0.051472265,
+              -0.01461242,
+              0.033237122,
+              0.023671487,
+              -0.022980422,
+              0.038988944,
+              0.030206418
+            ],
+            "index": 0,
+            "object": "embedding"
+          }
+        ],
+        "model": "all-minilm:l6-v2",
+        "object": "list",
+        "usage": {
+          "prompt_tokens": 2,
+          "total_tokens": 2
+        }
+      }
+    },
+    "is_streaming": false
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/batches/test_batches.py

@@ -323,3 +323,92 @@ class TestBatchesIntegration:
         if final_batch.error_file_id is not None:
             deleted_error_file = openai_client.files.delete(final_batch.error_file_id)
             assert deleted_error_file.deleted
+
+    def test_batch_e2e_embeddings(self, openai_client, batch_helper, embedding_model_id):
+        """Run an end-to-end batch with embeddings requests including both string and list inputs."""
+        batch_requests = [
+            {
+                "custom_id": "success-1",
+                "method": "POST",
+                "url": "/v1/embeddings",
+                "body": {"model": embedding_model_id, "input": "Hello world", "encoding_format": "float"},
+            },
+            {
+                "custom_id": "success-2",
+                "method": "POST",
+                "url": "/v1/embeddings",
+                "body": {
+                    "model": embedding_model_id,
+                    "input": ["How are you?", "Good morning", "Have a great day"],
+                    "encoding_format": "float",
+                },
+            },
+        ]
+
+        with batch_helper.create_file(batch_requests) as uploaded_file:
+            batch = openai_client.batches.create(
+                input_file_id=uploaded_file.id,
+                endpoint="/v1/embeddings",
+                completion_window="24h",
+                metadata={"test": "e2e_embeddings_success"},
+            )
+
+            final_batch = batch_helper.wait_for(
+                batch.id,
+                max_wait_time=3 * 60,
+                expected_statuses={"completed"},
+                timeout_action="skip",
+            )
+
+        assert final_batch.status == "completed"
+        assert final_batch.request_counts is not None
+        assert final_batch.request_counts.total == 2
+        assert final_batch.request_counts.completed == 2
+        assert final_batch.output_file_id is not None
+
+        output_content = openai_client.files.content(final_batch.output_file_id)
+        if isinstance(output_content, str):
+            output_text = output_content
+        else:
+            output_text = output_content.content.decode("utf-8")
+
+        output_lines = output_text.strip().split("\n")
+        assert len(output_lines) == 2
+
+        # Check first result (string input)
+        result1 = json.loads(output_lines[0])
+        assert result1["custom_id"] in ["success-1", "success-2"]
+        assert "response" in result1
+        assert result1["response"]["status_code"] == 200
+
+        # Verify the response body contains embeddings data
+        response_body1 = json.loads(result1["response"]["body"])
+        assert response_body1["object"] == "list"
+        assert "data" in response_body1
+        assert len(response_body1["data"]) == 1
+        assert "embedding" in response_body1["data"][0]
+        assert "index" in response_body1["data"][0]
+        assert response_body1["data"][0]["index"] == 0
+
+        # Check second result (list input)
+        result2 = json.loads(output_lines[1])
+        assert result2["custom_id"] in ["success-1", "success-2"]
+        assert "response" in result2
+        assert result2["response"]["status_code"] == 200
+
+        # Verify the response body contains embeddings data for list input
+        response_body2 = json.loads(result2["response"]["body"])
+        assert response_body2["object"] == "list"
+        assert "data" in response_body2
+        assert len(response_body2["data"]) == 3  # Three strings in the list
+        for i, embedding_data in enumerate(response_body2["data"]):
+            assert "embedding" in embedding_data
+            assert "index" in embedding_data
+            assert embedding_data["index"] == i
+
+        deleted_output_file = openai_client.files.delete(final_batch.output_file_id)
+        assert deleted_output_file.deleted
+
+        if final_batch.error_file_id is not None:
+            deleted_error_file = openai_client.files.delete(final_batch.error_file_id)
+            assert deleted_error_file.deleted

tests/integration/common/recordings/02c93bb3c314427bae2b7a7a6f054792b9f22d2cb4522eab802810be8672d3dc.json

@@ -12,28 +12,10 @@
     "body": {
       "__type__": "ollama._types.ProcessResponse",
       "__data__": {
-        "models": [
-          {
-            "model": "llama3.2-vision:11b",
-            "name": "llama3.2-vision:11b",
-            "digest": "6f2f9757ae97e8a3f8ea33d6adb2b11d93d9a35bef277cd2c0b1b5af8e8d0b1e",
-            "expires_at": "2025-10-08T12:40:47.430429-07:00",
-            "size": 11765236384,
-            "size_vram": 11765236384,
-            "details": {
-              "parent_model": "",
-              "format": "gguf",
-              "family": "mllama",
-              "families": [
-                "mllama"
-              ],
-              "parameter_size": "10.7B",
-              "quantization_level": "Q4_K_M"
-            }
-          }
-        ]
+        "models": []
       }
     },
     "is_streaming": false
-  }
+  },
+  "id_normalization_mapping": {}
 }

tests/integration/common/recordings/models-64a2277c90f0f42576f60c1030e3a020403d34a95f56931b792d5939f4cebc57-e660ee4a.json

@@ -93,5 +93,6 @@
       }
     ],
     "is_streaming": false
-  }
+  },
+  "id_normalization_mapping": {}
 }

tests/integration/responses/recordings/56a01bed56ba10824a14930eda83792ad086e12affe4c5f21f268f714d1fa41d.json

@@ -0,0 +1,273 @@
+{
+  "test_id": "tests/integration/responses/test_basic_responses.py::test_response_streaming_basic[client_with_models-txt=openai/gpt-4o-earth]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "Which planet do humans live on?"
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      }
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-56a01bed56ba",
+          "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_1827dd0c55",
+          "usage": null,
+          "obfuscation": "hA9E1EFNDdppk3"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-56a01bed56ba",
+          "choices": [
+            {
+              "delta": {
+                "content": "Hum",
+                "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_1827dd0c55",
+          "usage": null,
+          "obfuscation": "waalMD9oTwIyd"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-56a01bed56ba",
+          "choices": [
+            {
+              "delta": {
+                "content": "ans",
+                "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_1827dd0c55",
+          "usage": null,
+          "obfuscation": "vyTLO6JSApar2"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-56a01bed56ba",
+          "choices": [
+            {
+              "delta": {
+                "content": " live",
+                "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_1827dd0c55",
+          "usage": null,
+          "obfuscation": "KQU8fOkg5R6"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-56a01bed56ba",
+          "choices": [
+            {
+              "delta": {
+                "content": " on",
+                "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_1827dd0c55",
+          "usage": null,
+          "obfuscation": "j7D1acmtXVxh1"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-56a01bed56ba",
+          "choices": [
+            {
+              "delta": {
+                "content": " Earth",
+                "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_1827dd0c55",
+          "usage": null,
+          "obfuscation": "F4dwv7pOew"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-56a01bed56ba",
+          "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_1827dd0c55",
+          "usage": null,
+          "obfuscation": "4xPS4mauTagzkxj"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-56a01bed56ba",
+          "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_1827dd0c55",
+          "usage": null,
+          "obfuscation": "pjCVonxRXx"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-56a01bed56ba",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_1827dd0c55",
+          "usage": {
+            "completion_tokens": 6,
+            "prompt_tokens": 14,
+            "total_tokens": 20,
+            "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": ""
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/recordings/5a1a11d0b4987b7e5f654a3a22149616ff9261866ca6d343d1482723e9c38c81.json

@@ -0,0 +1,948 @@
+{
+  "test_id": "tests/integration/responses/test_basic_responses.py::test_response_streaming_basic[client_with_models-txt=openai/gpt-4o-saturn]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "Which planet has rings around it with a name starting with letter S?"
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      }
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "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": "UKZMfDO4alqdUr"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "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": "PauvZTS9JhHJQ"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "choices": [
+            {
+              "delta": {
+                "content": " planet",
+                "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": "I1PMKwlor"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "choices": [
+            {
+              "delta": {
+                "content": " with",
+                "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": "NKus7N8SshY"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "choices": [
+            {
+              "delta": {
+                "content": " rings",
+                "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": "hjPcORfLsk"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "choices": [
+            {
+              "delta": {
+                "content": " around",
+                "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": "ceeQx9iTw"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "choices": [
+            {
+              "delta": {
+                "content": " it",
+                "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": "8ad9CC6rmF939"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "choices": [
+            {
+              "delta": {
+                "content": " and",
+                "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": "9zCNPfknf9Wk"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "choices": [
+            {
+              "delta": {
+                "content": " a",
+                "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": "d0Aa4FFslxFL5t"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "choices": [
+            {
+              "delta": {
+                "content": " name",
+                "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": "xWj2hnWGsbL"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "choices": [
+            {
+              "delta": {
+                "content": " starting",
+                "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": "1pyuTcJ"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "choices": [
+            {
+              "delta": {
+                "content": " with",
+                "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": "l7uhYGpUcE1"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "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": "aj5jkF7RI6jc"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "choices": [
+            {
+              "delta": {
+                "content": " letter",
+                "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": "1Hqbz5J84"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "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": "IpvA6bGSmegM9"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "choices": [
+            {
+              "delta": {
+                "content": "S",
+                "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": "EzbeTEiigNgr1WP"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "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": "SpfVHvAhaLPtgf"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "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": "Tz0BHfbZzwIU8"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "choices": [
+            {
+              "delta": {
+                "content": " Saturn",
+                "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": "v62Yi0Uy5"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "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": "FaYqJrpzivGOEw6"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "choices": [
+            {
+              "delta": {
+                "content": " Saturn",
+                "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": "ni5aiHptn"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "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": "2Xbu8coUEl3nE"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "choices": [
+            {
+              "delta": {
+                "content": " well",
+                "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": "nOPBJSLkxK9"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "choices": [
+            {
+              "delta": {
+                "content": "-known",
+                "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": "65uX75JvoU"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "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": "RgOnunwbvzQQ"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "choices": [
+            {
+              "delta": {
+                "content": " its",
+                "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": "s9LwMsjnLC8q"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "choices": [
+            {
+              "delta": {
+                "content": " prominent",
+                "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": "mPdj0V"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "choices": [
+            {
+              "delta": {
+                "content": " and",
+                "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": "hvjs7fxAOSUX"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "choices": [
+            {
+              "delta": {
+                "content": " beautiful",
+                "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": "WVtGRs"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "choices": [
+            {
+              "delta": {
+                "content": " ring",
+                "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": "B0ahJwERcG6"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "choices": [
+            {
+              "delta": {
+                "content": " system",
+                "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": "PRkj9JNVV"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "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": "cvQVcsFDMfrfnVh"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "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": "kBSMtrCs6N"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-5a1a11d0b498",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": {
+            "completion_tokens": 31,
+            "prompt_tokens": 21,
+            "total_tokens": 52,
+            "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": "QRz5sPimoUuAGY0"
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/recordings/792f5c2603ba2d22d48aebd47738190504316f2b7821673677c4af11131b4f0b.json

@@ -0,0 +1,335 @@
+{
+  "test_id": "tests/integration/responses/test_basic_responses.py::test_response_non_streaming_basic[client_with_models-txt=openai/gpt-4o-earth]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "Which planet do humans live on?"
+        },
+        {
+          "role": "assistant",
+          "content": "Humans live on Earth."
+        },
+        {
+          "role": "user",
+          "content": "Repeat your previous response in all caps."
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      }
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-792f5c2603ba",
+          "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_f33640a400",
+          "usage": null,
+          "obfuscation": "Bx83B946ygH7Vb"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-792f5c2603ba",
+          "choices": [
+            {
+              "delta": {
+                "content": "H",
+                "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_f33640a400",
+          "usage": null,
+          "obfuscation": "jiQelGI7oxHigEi"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-792f5c2603ba",
+          "choices": [
+            {
+              "delta": {
+                "content": "UM",
+                "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_f33640a400",
+          "usage": null,
+          "obfuscation": "ePc4EP3mEfA9JN"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-792f5c2603ba",
+          "choices": [
+            {
+              "delta": {
+                "content": "ANS",
+                "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_f33640a400",
+          "usage": null,
+          "obfuscation": "CSsCO4DtCo1jC"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-792f5c2603ba",
+          "choices": [
+            {
+              "delta": {
+                "content": " LIVE",
+                "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_f33640a400",
+          "usage": null,
+          "obfuscation": "uBbrt1MQwxP"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-792f5c2603ba",
+          "choices": [
+            {
+              "delta": {
+                "content": " ON",
+                "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_f33640a400",
+          "usage": null,
+          "obfuscation": "0UHMl6xkx8nyl"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-792f5c2603ba",
+          "choices": [
+            {
+              "delta": {
+                "content": " EAR",
+                "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_f33640a400",
+          "usage": null,
+          "obfuscation": "yy5k0fpmcCxH"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-792f5c2603ba",
+          "choices": [
+            {
+              "delta": {
+                "content": "TH",
+                "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_f33640a400",
+          "usage": null,
+          "obfuscation": "actry6zH88X5t8"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-792f5c2603ba",
+          "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_f33640a400",
+          "usage": null,
+          "obfuscation": "5FuIwYGrTdqVBbu"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-792f5c2603ba",
+          "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_f33640a400",
+          "usage": null,
+          "obfuscation": "kHtcUGMzub"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-792f5c2603ba",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_f33640a400",
+          "usage": {
+            "completion_tokens": 8,
+            "prompt_tokens": 36,
+            "total_tokens": 44,
+            "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": ""
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/recordings/8dd4f0615edfc30423d90b13daac2116ef34b6d673c83743a9e22a2b56f08eb2.json

@@ -41,9 +41,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_cbf1785567",
+          "system_fingerprint": "fp_f33640a400",
           "usage": null,
-          "obfuscation": "zftbFaqu4Bi"
+          "obfuscation": "eQtIB22UP5y"
         }
       },
       {
@@ -68,9 +68,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_cbf1785567",
+          "system_fingerprint": "fp_f33640a400",
           "usage": null,
-          "obfuscation": "ItcaW7b3D9"
+          "obfuscation": "xVnzycGrGY"
         }
       },
       {
@@ -95,9 +95,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_cbf1785567",
+          "system_fingerprint": "fp_f33640a400",
           "usage": null,
-          "obfuscation": "n9CzCrtHvG"
+          "obfuscation": "iCxzQ9UFRl"
         }
       },
       {
@@ -122,9 +122,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_cbf1785567",
+          "system_fingerprint": "fp_f33640a400",
           "usage": null,
-          "obfuscation": "fWwHGFYy"
+          "obfuscation": "lVzetb7n"
         }
       },
       {
@@ -149,9 +149,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_cbf1785567",
+          "system_fingerprint": "fp_f33640a400",
           "usage": null,
-          "obfuscation": "kdNkbMvQY5"
+          "obfuscation": "Oi2xzNkyQQ"
         }
       },
       {
@@ -176,9 +176,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_cbf1785567",
+          "system_fingerprint": "fp_f33640a400",
           "usage": null,
-          "obfuscation": "DhWaK87"
+          "obfuscation": "SBQnBFp"
         }
       },
       {
@@ -203,9 +203,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_cbf1785567",
+          "system_fingerprint": "fp_f33640a400",
           "usage": null,
-          "obfuscation": "ocA1GIrZyPFW"
+          "obfuscation": "CtftBIpiBh2x"
         }
       },
       {
@@ -230,12 +230,13 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_cbf1785567",
+          "system_fingerprint": "fp_f33640a400",
           "usage": null,
-          "obfuscation": "jlCQlW0"
+          "obfuscation": "xN1HF2a"
         }
       }
     ],
     "is_streaming": true
-  }
+  },
+  "id_normalization_mapping": {}
 }

tests/integration/responses/recordings/cc0156eadb588e2cdfb9d4d0db0355b6a1d6c57e467f136f45586b73bf95766d.json

@@ -41,9 +41,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_f33640a400",
+          "system_fingerprint": "fp_1827dd0c55",
           "usage": null,
-          "obfuscation": "zqLiclPltu3"
+          "obfuscation": "wl1L4ubkZC2"
         }
       },
       {
@@ -68,9 +68,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_f33640a400",
+          "system_fingerprint": "fp_1827dd0c55",
           "usage": null,
-          "obfuscation": "grzpYaHPM8"
+          "obfuscation": "MKXxbHEqGM"
         }
       },
       {
@@ -95,9 +95,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_f33640a400",
+          "system_fingerprint": "fp_1827dd0c55",
           "usage": null,
-          "obfuscation": "zEX86j"
+          "obfuscation": "RXhSL9"
         }
       },
       {
@@ -122,9 +122,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_f33640a400",
+          "system_fingerprint": "fp_1827dd0c55",
           "usage": null,
-          "obfuscation": "fvu80DBX"
+          "obfuscation": "RTRK7mJP"
         }
       },
       {
@@ -149,9 +149,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_f33640a400",
+          "system_fingerprint": "fp_1827dd0c55",
           "usage": null,
-          "obfuscation": "iIA0sN8"
+          "obfuscation": "6EhZvCD"
         }
       },
       {
@@ -176,9 +176,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_f33640a400",
+          "system_fingerprint": "fp_1827dd0c55",
           "usage": null,
-          "obfuscation": "fohE4v"
+          "obfuscation": "qiNxmx"
         }
       },
       {
@@ -203,9 +203,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_f33640a400",
+          "system_fingerprint": "fp_1827dd0c55",
           "usage": null,
-          "obfuscation": "9jaVkFkWKk"
+          "obfuscation": "RNXYcBmW82"
         }
       },
       {
@@ -215,7 +215,7 @@
           "choices": [
             {
               "delta": {
-                "content": " and",
+                "content": ",",
                 "function_call": null,
                 "refusal": null,
                 "role": null,
@@ -230,9 +230,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_f33640a400",
+          "system_fingerprint": "fp_1827dd0c55",
           "usage": null,
-          "obfuscation": "4XV7CDgbt"
+          "obfuscation": "dRGyq9aos4Ao"
         }
       },
       {
@@ -242,7 +242,7 @@
           "choices": [
             {
               "delta": {
-                "content": " a",
+                "content": " whose",
                 "function_call": null,
                 "refusal": null,
                 "role": null,
@@ -257,9 +257,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_f33640a400",
+          "system_fingerprint": "fp_1827dd0c55",
           "usage": null,
-          "obfuscation": "8q9WqFBh4Y0"
+          "obfuscation": "iHNHPzN"
         }
       },
       {
@@ -284,9 +284,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_f33640a400",
+          "system_fingerprint": "fp_1827dd0c55",
           "usage": null,
-          "obfuscation": "wUB8pf8C"
+          "obfuscation": "CWIF5di5"
         }
       },
       {
@@ -296,7 +296,7 @@
           "choices": [
             {
               "delta": {
-                "content": " starting",
+                "content": " starts",
                 "function_call": null,
                 "refusal": null,
                 "role": null,
@@ -311,9 +311,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_f33640a400",
+          "system_fingerprint": "fp_1827dd0c55",
           "usage": null,
-          "obfuscation": "TX3B"
+          "obfuscation": "TQurN8"
         }
       },
       {
@@ -338,9 +338,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_f33640a400",
+          "system_fingerprint": "fp_1827dd0c55",
           "usage": null,
-          "obfuscation": "KuxjVC7G"
+          "obfuscation": "NcKx1q81"
         }
       },
       {
@@ -365,9 +365,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_f33640a400",
+          "system_fingerprint": "fp_1827dd0c55",
           "usage": null,
-          "obfuscation": "dX94lMG2M"
+          "obfuscation": "EQ0qBHCjl"
         }
       },
       {
@@ -392,9 +392,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_f33640a400",
+          "system_fingerprint": "fp_1827dd0c55",
           "usage": null,
-          "obfuscation": "N6f05V"
+          "obfuscation": "cjvz4i"
         }
       },
       {
@@ -419,9 +419,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_f33640a400",
+          "system_fingerprint": "fp_1827dd0c55",
           "usage": null,
-          "obfuscation": "Gf1gjnyVAI"
+          "obfuscation": "brfeAXs8FX"
         }
       },
       {
@@ -446,9 +446,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_f33640a400",
+          "system_fingerprint": "fp_1827dd0c55",
           "usage": null,
-          "obfuscation": "TQ2IiLZtvxzh"
+          "obfuscation": "RQJHhiH9GkeD"
         }
       },
       {
@@ -458,7 +458,7 @@
           "choices": [
             {
               "delta": {
-                "content": "\"",
+                "content": ",\"",
                 "function_call": null,
                 "refusal": null,
                 "role": null,
@@ -473,9 +473,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_f33640a400",
+          "system_fingerprint": "fp_1827dd0c55",
           "usage": null,
-          "obfuscation": "aHvCedShvKO"
+          "obfuscation": "ZF32Jc9E2G"
         }
       },
       {
@@ -500,9 +500,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_f33640a400",
+          "system_fingerprint": "fp_1827dd0c55",
           "usage": null,
-          "obfuscation": "fhDv1M3NuQ"
+          "obfuscation": "G2odltRd0L"
         }
       },
       {
@@ -527,9 +527,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_f33640a400",
+          "system_fingerprint": "fp_1827dd0c55",
           "usage": null,
-          "obfuscation": "xwJPIP"
+          "obfuscation": "2Osmry"
         }
       },
       {
@@ -554,9 +554,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_f33640a400",
+          "system_fingerprint": "fp_1827dd0c55",
           "usage": null,
-          "obfuscation": "75QYRcPV0IuE"
+          "obfuscation": "7SNbKabEykCr"
         }
       },
       {
@@ -581,9 +581,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_f33640a400",
+          "system_fingerprint": "fp_1827dd0c55",
           "usage": null,
-          "obfuscation": "A4Rh3S"
+          "obfuscation": "H350qD"
         }
       },
       {
@@ -608,9 +608,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_f33640a400",
+          "system_fingerprint": "fp_1827dd0c55",
           "usage": null,
-          "obfuscation": "Bohe93xwDk"
+          "obfuscation": "YJTtTjmwSX"
         }
       },
       {
@@ -635,9 +635,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_f33640a400",
+          "system_fingerprint": "fp_1827dd0c55",
           "usage": null,
-          "obfuscation": "6f77oxEV"
+          "obfuscation": "73nC8tFe"
         }
       },
       {
@@ -662,9 +662,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_f33640a400",
+          "system_fingerprint": "fp_1827dd0c55",
           "usage": null,
-          "obfuscation": "RCyK7L4"
+          "obfuscation": "nHEeUa9"
         }
       },
       {
@@ -689,9 +689,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_f33640a400",
+          "system_fingerprint": "fp_1827dd0c55",
           "usage": null,
-          "obfuscation": "s8cnTAnao"
+          "obfuscation": "9W2IOivfJ"
         }
       },
       {
@@ -716,9 +716,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_f33640a400",
+          "system_fingerprint": "fp_1827dd0c55",
           "usage": null,
-          "obfuscation": "uI9KpovPR"
+          "obfuscation": "WlCJ1unfP"
         }
       },
       {
@@ -743,9 +743,63 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_f33640a400",
+          "system_fingerprint": "fp_1827dd0c55",
           "usage": null,
-          "obfuscation": "pWL"
+          "obfuscation": "LY2"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-cc0156eadb58",
+          "choices": [
+            {
+              "delta": {
+                "content": " and",
+                "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_1827dd0c55",
+          "usage": null,
+          "obfuscation": "vw5KZwo9J"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-cc0156eadb58",
+          "choices": [
+            {
+              "delta": {
+                "content": " complex",
+                "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_1827dd0c55",
+          "usage": null,
+          "obfuscation": "YUfqU"
         }
       },
       {
@@ -770,9 +824,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_f33640a400",
+          "system_fingerprint": "fp_1827dd0c55",
           "usage": null,
-          "obfuscation": "jupSPBuU"
+          "obfuscation": "WAHNZQCW"
         }
       },
       {
@@ -797,9 +851,333 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_f33640a400",
+          "system_fingerprint": "fp_1827dd0c55",
           "usage": null,
-          "obfuscation": "ZZ5e4f"
+          "obfuscation": "iCsrlh"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-cc0156eadb58",
+          "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_1827dd0c55",
+          "usage": null,
+          "obfuscation": "OssepLC4D5xV"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-cc0156eadb58",
+          "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_1827dd0c55",
+          "usage": null,
+          "obfuscation": "KYxhInr"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-cc0156eadb58",
+          "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_1827dd0c55",
+          "usage": null,
+          "obfuscation": "vsMjIHuV1f"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-cc0156eadb58",
+          "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_1827dd0c55",
+          "usage": null,
+          "obfuscation": "a4hilLjNF"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-cc0156eadb58",
+          "choices": [
+            {
+              "delta": {
+                "content": " most",
+                "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_1827dd0c55",
+          "usage": null,
+          "obfuscation": "PSmGYjCZ"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-cc0156eadb58",
+          "choices": [
+            {
+              "delta": {
+                "content": " extensive",
+                "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_1827dd0c55",
+          "usage": null,
+          "obfuscation": "sdu"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-cc0156eadb58",
+          "choices": [
+            {
+              "delta": {
+                "content": " and",
+                "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_1827dd0c55",
+          "usage": null,
+          "obfuscation": "ruUKcQ2AO"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-cc0156eadb58",
+          "choices": [
+            {
+              "delta": {
+                "content": " visible",
+                "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_1827dd0c55",
+          "usage": null,
+          "obfuscation": "G6PXg"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-cc0156eadb58",
+          "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_1827dd0c55",
+          "usage": null,
+          "obfuscation": "uYj8td6Yf5"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-cc0156eadb58",
+          "choices": [
+            {
+              "delta": {
+                "content": " our",
+                "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_1827dd0c55",
+          "usage": null,
+          "obfuscation": "jEDd7GvP7"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-cc0156eadb58",
+          "choices": [
+            {
+              "delta": {
+                "content": " solar",
+                "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_1827dd0c55",
+          "usage": null,
+          "obfuscation": "crXMLcu"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-cc0156eadb58",
+          "choices": [
+            {
+              "delta": {
+                "content": " system",
+                "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_1827dd0c55",
+          "usage": null,
+          "obfuscation": "8wRexB"
         }
       },
       {
@@ -824,9 +1202,9 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_f33640a400",
+          "system_fingerprint": "fp_1827dd0c55",
           "usage": null,
-          "obfuscation": "GyHY0sipKZ24"
+          "obfuscation": "VQpIxgZRp9bP"
         }
       },
       {
@@ -851,12 +1229,13 @@
           "model": "gpt-4o-2024-08-06",
           "object": "chat.completion.chunk",
           "service_tier": "default",
-          "system_fingerprint": "fp_f33640a400",
+          "system_fingerprint": "fp_1827dd0c55",
           "usage": null,
-          "obfuscation": "gMrZjfo"
+          "obfuscation": "1PL6rSg"
         }
       }
     ],
     "is_streaming": true
-  }
+  },
+  "id_normalization_mapping": {}
 }

tests/integration/responses/recordings/d058ce92a083c160c08eb29e7b396e65fa55f4ddf217d2bd3c92eeb85bee5251.json

@@ -0,0 +1,894 @@
+{
+  "test_id": "tests/integration/responses/test_basic_responses.py::test_response_non_streaming_basic[client_with_models-txt=openai/gpt-4o-saturn]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "Which planet has rings around it with a name starting with letter S?"
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      }
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d058ce92a083",
+          "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": "z56q5oZHGZKdtw"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d058ce92a083",
+          "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": "LmzSBsnBqMCfY"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d058ce92a083",
+          "choices": [
+            {
+              "delta": {
+                "content": " planet",
+                "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": "0cyvj0Txx"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d058ce92a083",
+          "choices": [
+            {
+              "delta": {
+                "content": " with",
+                "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": "7NDFvRwCV4H"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d058ce92a083",
+          "choices": [
+            {
+              "delta": {
+                "content": " rings",
+                "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": "cXFnChS2le"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d058ce92a083",
+          "choices": [
+            {
+              "delta": {
+                "content": " around",
+                "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": "KjBEmV3kP"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d058ce92a083",
+          "choices": [
+            {
+              "delta": {
+                "content": " it",
+                "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": "HRuZPETdawxRE"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d058ce92a083",
+          "choices": [
+            {
+              "delta": {
+                "content": " that",
+                "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": "oWgRcLBEtgJ"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d058ce92a083",
+          "choices": [
+            {
+              "delta": {
+                "content": " starts",
+                "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": "57Wb5VAIC"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d058ce92a083",
+          "choices": [
+            {
+              "delta": {
+                "content": " with",
+                "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": "PIJlSXUHBpa"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d058ce92a083",
+          "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": "z1j4e36eJaU4"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d058ce92a083",
+          "choices": [
+            {
+              "delta": {
+                "content": " letter",
+                "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": "jDq6ugLXW"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d058ce92a083",
+          "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": "Tl7ZVn6xaBP3M"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d058ce92a083",
+          "choices": [
+            {
+              "delta": {
+                "content": "S",
+                "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": "Jgd6CpJfjXpHnNW"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d058ce92a083",
+          "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": "OYepkOsMHU8jw5"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d058ce92a083",
+          "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": "6UnjEo5YWE1zb"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d058ce92a083",
+          "choices": [
+            {
+              "delta": {
+                "content": " Saturn",
+                "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": "ljnq8v7G7"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d058ce92a083",
+          "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": "d79gDV0DqNlOO51"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d058ce92a083",
+          "choices": [
+            {
+              "delta": {
+                "content": " Saturn",
+                "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": "bwM1CRrwC"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d058ce92a083",
+          "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": "uTygnf5A75n5i"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d058ce92a083",
+          "choices": [
+            {
+              "delta": {
+                "content": " well",
+                "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": "J2mPiUrRBrP"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d058ce92a083",
+          "choices": [
+            {
+              "delta": {
+                "content": " known",
+                "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": "ySgqipG0Qe"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d058ce92a083",
+          "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": "KiyTURpCBdJM"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d058ce92a083",
+          "choices": [
+            {
+              "delta": {
+                "content": " its",
+                "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": "JBRXJQhWdglI"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d058ce92a083",
+          "choices": [
+            {
+              "delta": {
+                "content": " prominent",
+                "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": "0mjYkG"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d058ce92a083",
+          "choices": [
+            {
+              "delta": {
+                "content": " and",
+                "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": "hR3bcUfNkRTx"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d058ce92a083",
+          "choices": [
+            {
+              "delta": {
+                "content": " extensive",
+                "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": "18tm6k"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d058ce92a083",
+          "choices": [
+            {
+              "delta": {
+                "content": " ring",
+                "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": "ZIL4ZROICo4"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d058ce92a083",
+          "choices": [
+            {
+              "delta": {
+                "content": " system",
+                "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": "EVoEpQkJL"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d058ce92a083",
+          "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": "cE7oSZixhWV42Kn"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d058ce92a083",
+          "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": "Dl1rR45BAF"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-d058ce92a083",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": {
+            "completion_tokens": 29,
+            "prompt_tokens": 21,
+            "total_tokens": 50,
+            "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": "FXirZrSNkvTrASR"
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/recordings/dca42c8244f72a79450c018f63c234d8026087d4416c90a082bd6efb53d00f11.json

@@ -0,0 +1,273 @@
+{
+  "test_id": "tests/integration/responses/test_basic_responses.py::test_response_non_streaming_basic[client_with_models-txt=openai/gpt-4o-earth]",
+  "request": {
+    "method": "POST",
+    "url": "https://api.openai.com/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "gpt-4o",
+      "messages": [
+        {
+          "role": "user",
+          "content": "Which planet do humans live on?"
+        }
+      ],
+      "stream": true,
+      "stream_options": {
+        "include_usage": true
+      }
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "gpt-4o"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-dca42c8244f7",
+          "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": "uWM4CTrTBdhbyI"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-dca42c8244f7",
+          "choices": [
+            {
+              "delta": {
+                "content": "Hum",
+                "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": "JuK3luIjXIAR0"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-dca42c8244f7",
+          "choices": [
+            {
+              "delta": {
+                "content": "ans",
+                "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": "mKYRNnis1mCZ4"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-dca42c8244f7",
+          "choices": [
+            {
+              "delta": {
+                "content": " live",
+                "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": "SrAIMLQQ8Gj"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-dca42c8244f7",
+          "choices": [
+            {
+              "delta": {
+                "content": " on",
+                "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": "2fzicdQg1s2C8"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-dca42c8244f7",
+          "choices": [
+            {
+              "delta": {
+                "content": " Earth",
+                "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": "saIKgW0kJK"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-dca42c8244f7",
+          "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": "y4Bq9fWigLsde0J"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-dca42c8244f7",
+          "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": "5VaLcoYLcc"
+        }
+      },
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-dca42c8244f7",
+          "choices": [],
+          "created": 0,
+          "model": "gpt-4o-2024-08-06",
+          "object": "chat.completion.chunk",
+          "service_tier": "default",
+          "system_fingerprint": "fp_cbf1785567",
+          "usage": {
+            "completion_tokens": 6,
+            "prompt_tokens": 14,
+            "total_tokens": 20,
+            "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": ""
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/responses/recordings/f06eb8bec25da44281fda0ff56610393b3c19a3c7fe28b4f773dcf0af15b0c99.json

@@ -43,7 +43,7 @@
           "service_tier": "default",
           "system_fingerprint": "fp_f33640a400",
           "usage": null,
-          "obfuscation": "zz0dWhiq9Lc"
+          "obfuscation": "zV8s7dKVA0e"
         }
       },
       {
@@ -70,7 +70,7 @@
           "service_tier": "default",
           "system_fingerprint": "fp_f33640a400",
           "usage": null,
-          "obfuscation": "apZyVste9l"
+          "obfuscation": "DoxBbL71uu"
         }
       },
       {
@@ -97,7 +97,7 @@
           "service_tier": "default",
           "system_fingerprint": "fp_f33640a400",
           "usage": null,
-          "obfuscation": "bKzudmyiZd"
+          "obfuscation": "3kPw1tJtBy"
         }
       },
       {
@@ -124,7 +124,7 @@
           "service_tier": "default",
           "system_fingerprint": "fp_f33640a400",
           "usage": null,
-          "obfuscation": "OgfYmGnf"
+          "obfuscation": "GYMVbcCI"
         }
       },
       {
@@ -151,7 +151,7 @@
           "service_tier": "default",
           "system_fingerprint": "fp_f33640a400",
           "usage": null,
-          "obfuscation": "nWM9P08zaL"
+          "obfuscation": "0alfiwg16c"
         }
       },
       {
@@ -178,7 +178,7 @@
           "service_tier": "default",
           "system_fingerprint": "fp_f33640a400",
           "usage": null,
-          "obfuscation": "gno9CMj"
+          "obfuscation": "Wug4Phl"
         }
       },
       {
@@ -205,7 +205,7 @@
           "service_tier": "default",
           "system_fingerprint": "fp_f33640a400",
           "usage": null,
-          "obfuscation": "iheIwJKTdAVV"
+          "obfuscation": "cIZ0eTIUOj4x"
         }
       },
       {
@@ -232,10 +232,11 @@
           "service_tier": "default",
           "system_fingerprint": "fp_f33640a400",
           "usage": null,
-          "obfuscation": "X6N8Z47"
+          "obfuscation": "FbdywkT"
         }
       }
     ],
     "is_streaming": true
-  }
+  },
+  "id_normalization_mapping": {}
 }

tests/integration/responses/test_basic_responses.py

@@ -23,6 +23,14 @@ def test_response_non_streaming_basic(compat_client, text_model_id, case):
     assert len(output_text) > 0
     assert case.expected.lower() in output_text
 
+    # Verify usage is reported
+    assert response.usage is not None, "Response should include usage information"
+    assert response.usage.input_tokens > 0, "Input tokens should be greater than 0"
+    assert response.usage.output_tokens > 0, "Output tokens should be greater than 0"
+    assert response.usage.total_tokens == response.usage.input_tokens + response.usage.output_tokens, (
+        "Total tokens should equal input + output tokens"
+    )
+
     retrieved_response = compat_client.responses.retrieve(response_id=response.id)
     assert retrieved_response.output_text == response.output_text
 
@@ -73,6 +81,15 @@ def test_response_streaming_basic(compat_client, text_model_id, case):
             assert len(output_text) > 0, "Response should have content"
             assert case.expected.lower() in output_text, f"Expected '{case.expected}' in response"
 
+            # Verify usage is reported in final response
+            assert chunk.response.usage is not None, "Completed response should include usage information"
+            assert chunk.response.usage.input_tokens > 0, "Input tokens should be greater than 0"
+            assert chunk.response.usage.output_tokens > 0, "Output tokens should be greater than 0"
+            assert (
+                chunk.response.usage.total_tokens
+                == chunk.response.usage.input_tokens + chunk.response.usage.output_tokens
+            ), "Total tokens should equal input + output tokens"
+
     # Use validator for common checks
     validator = StreamingValidator(events)
     validator.assert_basic_event_sequence()

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

@@ -175,6 +175,9 @@ async def test_create_openai_response_with_string_input(openai_responses_impl, m
         tools=None,
         stream=True,
         temperature=0.1,
+        stream_options={
+            "include_usage": True,
+        },
     )
 
     # Should have content part events for text streaming

tests/unit/providers/batches/test_reference.py

@@ -213,7 +213,6 @@ class TestReferenceBatchesImpl:
     @pytest.mark.parametrize(
         "endpoint",
         [
-            "/v1/embeddings",
             "/v1/invalid/endpoint",
             "",
         ],
@@ -765,3 +764,12 @@ class TestReferenceBatchesImpl:
         await asyncio.sleep(0.042)  # let tasks start
 
         assert active_batches == 2, f"Expected 2 active batches, got {active_batches}"
+
+    async def test_create_batch_embeddings_endpoint(self, provider):
+        """Test that batch creation succeeds with embeddings endpoint."""
+        batch = await provider.create_batch(
+            input_file_id="file_123",
+            endpoint="/v1/embeddings",
+            completion_window="24h",
+        )
+        assert batch.endpoint == "/v1/embeddings"

tests/unit/server/test_auth.py

@@ -122,7 +122,7 @@ def mock_impls():
 
 
 @pytest.fixture
-def scope_middleware_with_mocks(mock_auth_endpoint):
+def middleware_with_mocks(mock_auth_endpoint):
     """Create AuthenticationMiddleware with mocked route implementations"""
     mock_app = AsyncMock()
     auth_config = AuthenticationConfig(
@@ -137,18 +137,20 @@ def scope_middleware_with_mocks(mock_auth_endpoint):
     # Mock the route_impls to simulate finding routes with required scopes
     from llama_stack.schema_utils import WebMethod
 
-    scoped_webmethod = WebMethod(route="/test/scoped", method="POST", required_scope="test.read")
-
-    public_webmethod = WebMethod(route="/test/public", method="GET")
+    routes = {
+        ("POST", "/test/scoped"): WebMethod(route="/test/scoped", method="POST", required_scope="test.read"),
+        ("GET", "/test/public"): WebMethod(route="/test/public", method="GET"),
+        ("GET", "/health"): WebMethod(route="/health", method="GET", require_authentication=False),
+        ("GET", "/version"): WebMethod(route="/version", method="GET", require_authentication=False),
+        ("GET", "/models/list"): WebMethod(route="/models/list", method="GET", require_authentication=True),
+    }
 
     # Mock the route finding logic
     def mock_find_matching_route(method, path, route_impls):
-        if method == "POST" and path == "/test/scoped":
-            return None, {}, "/test/scoped", scoped_webmethod
-        elif method == "GET" and path == "/test/public":
-            return None, {}, "/test/public", public_webmethod
-        else:
-            raise ValueError("No matching route")
+        webmethod = routes.get((method, path))
+        if webmethod:
+            return None, {}, path, webmethod
+        raise ValueError("No matching route")
 
     import llama_stack.core.server.auth
 
@@ -659,9 +661,9 @@ def test_valid_introspection_with_custom_mapping_authentication(
 
 # Scope-based authorization tests
 @patch("httpx.AsyncClient.post", new=mock_post_success_with_scope)
-async def test_scope_authorization_success(scope_middleware_with_mocks, valid_api_key):
+async def test_scope_authorization_success(middleware_with_mocks, valid_api_key):
     """Test that user with required scope can access protected endpoint"""
-    middleware, mock_app = scope_middleware_with_mocks
+    middleware, mock_app = middleware_with_mocks
     mock_receive = AsyncMock()
     mock_send = AsyncMock()
 
@@ -680,9 +682,9 @@ async def test_scope_authorization_success(scope_middleware_with_mocks, valid_ap
 
 
 @patch("httpx.AsyncClient.post", new=mock_post_success_no_scope)
-async def test_scope_authorization_denied(scope_middleware_with_mocks, valid_api_key):
+async def test_scope_authorization_denied(middleware_with_mocks, valid_api_key):
     """Test that user without required scope gets 403 access denied"""
-    middleware, mock_app = scope_middleware_with_mocks
+    middleware, mock_app = middleware_with_mocks
     mock_receive = AsyncMock()
     mock_send = AsyncMock()
 
@@ -710,9 +712,9 @@ async def test_scope_authorization_denied(scope_middleware_with_mocks, valid_api
 
 
 @patch("httpx.AsyncClient.post", new=mock_post_success_no_scope)
-async def test_public_endpoint_no_scope_required(scope_middleware_with_mocks, valid_api_key):
+async def test_public_endpoint_no_scope_required(middleware_with_mocks, valid_api_key):
     """Test that public endpoints work without specific scopes"""
-    middleware, mock_app = scope_middleware_with_mocks
+    middleware, mock_app = middleware_with_mocks
     mock_receive = AsyncMock()
     mock_send = AsyncMock()
 
@@ -730,9 +732,9 @@ async def test_public_endpoint_no_scope_required(scope_middleware_with_mocks, va
     mock_send.assert_not_called()
 
 
-async def test_scope_authorization_no_auth_disabled(scope_middleware_with_mocks):
+async def test_scope_authorization_no_auth_disabled(middleware_with_mocks):
     """Test that when auth is disabled (no user), scope checks are bypassed"""
-    middleware, mock_app = scope_middleware_with_mocks
+    middleware, mock_app = middleware_with_mocks
     mock_receive = AsyncMock()
     mock_send = AsyncMock()
 
@@ -907,3 +909,41 @@ def test_kubernetes_auth_request_payload(kubernetes_auth_client, valid_token, mo
         request_body = call_args[1]["json"]
         assert request_body["apiVersion"] == "authentication.k8s.io/v1"
         assert request_body["kind"] == "SelfSubjectReview"
+
+
+async def test_unauthenticated_endpoint_access_health(middleware_with_mocks):
+    """Test that /health endpoints can be accessed without authentication"""
+    middleware, mock_app = middleware_with_mocks
+
+    # Test request to /health without auth header (level prefix v1 is added by router)
+    scope = {"type": "http", "path": "/health", "headers": [], "method": "GET"}
+    receive = AsyncMock()
+    send = AsyncMock()
+
+    # Should allow the request to proceed without authentication
+    await middleware(scope, receive, send)
+
+    # Verify that the request was passed to the app
+    mock_app.assert_called_once_with(scope, receive, send)
+
+    # Verify that no error response was sent
+    assert not any(call[0][0].get("status") == 401 for call in send.call_args_list)
+
+
+async def test_unauthenticated_endpoint_denied_for_other_paths(middleware_with_mocks):
+    """Test that endpoints other than /health and /version require authentication"""
+    middleware, mock_app = middleware_with_mocks
+
+    # Test request to /models/list without auth header
+    scope = {"type": "http", "path": "/models/list", "headers": [], "method": "GET"}
+    receive = AsyncMock()
+    send = AsyncMock()
+
+    # Should return 401 error
+    await middleware(scope, receive, send)
+
+    # Verify that the app was NOT called
+    mock_app.assert_not_called()
+
+    # Verify that a 401 error response was sent
+    assert any(call[0][0].get("status") == 401 for call in send.call_args_list)