feat: Add instructions parameter in response object (#3741)

# Problem
The current inline provider appends the user provided instructions to
messages as a system prompt, but the returned response object does not
contain the instructions field (as specified in the OpenAI responses
spec).

# What does this PR do?
This pull request adds the instruction field to the response object
definition and updates the inline provider. It also ensures that
instructions from previous response is not carried over to the next
response (as specified in the openAI spec).

Closes #[3566](https://github.com/llamastack/llama-stack/issues/3566)

## Test Plan

- Tested manually for change in model response w.r.t supplied
instructions field.
- Added unit test to check that the instructions from previous response
is not carried over to the next response.
- Added integration tests to check instructions parameter in the
returned response object.
- Added new recordings for the integration tests.

---------

Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>

docs/static/deprecated-llama-stack-spec.html

@@ -9024,6 +9024,10 @@
                         "$ref": "#/components/schemas/OpenAIResponseUsage",
                         "description": "(Optional) Token usage information for the response"
                     },
+                    "instructions": {
+                        "type": "string",
+                        "description": "(Optional) System message inserted into the model's context"
+                    },
                     "input": {
                         "type": "array",
                         "items": {
@@ -9901,6 +9905,10 @@
                     "usage": {
                         "$ref": "#/components/schemas/OpenAIResponseUsage",
                         "description": "(Optional) Token usage information for the response"
+                    },
+                    "instructions": {
+                        "type": "string",
+                        "description": "(Optional) System message inserted into the model's context"
                     }
                 },
                 "additionalProperties": false,

docs/static/deprecated-llama-stack-spec.yaml

@@ -6734,6 +6734,10 @@ components:
           $ref: '#/components/schemas/OpenAIResponseUsage'
           description: >-
             (Optional) Token usage information for the response
+        instructions:
+          type: string
+          description: >-
+            (Optional) System message inserted into the model's context
         input:
           type: array
           items:
@@ -7403,6 +7407,10 @@ components:
           $ref: '#/components/schemas/OpenAIResponseUsage'
           description: >-
             (Optional) Token usage information for the response
+        instructions:
+          type: string
+          description: >-
+            (Optional) System message inserted into the model's context
       additionalProperties: false
       required:
         - created_at

docs/static/llama-stack-spec.html

@@ -7600,6 +7600,10 @@
                         "$ref": "#/components/schemas/OpenAIResponseUsage",
                         "description": "(Optional) Token usage information for the response"
                     },
+                    "instructions": {
+                        "type": "string",
+                        "description": "(Optional) System message inserted into the model's context"
+                    },
                     "input": {
                         "type": "array",
                         "items": {
@@ -8148,6 +8152,10 @@
                     "usage": {
                         "$ref": "#/components/schemas/OpenAIResponseUsage",
                         "description": "(Optional) Token usage information for the response"
+                    },
+                    "instructions": {
+                        "type": "string",
+                        "description": "(Optional) System message inserted into the model's context"
                     }
                 },
                 "additionalProperties": false,

docs/static/llama-stack-spec.yaml

@@ -5815,6 +5815,10 @@ components:
           $ref: '#/components/schemas/OpenAIResponseUsage'
           description: >-
             (Optional) Token usage information for the response
+        instructions:
+          type: string
+          description: >-
+            (Optional) System message inserted into the model's context
         input:
           type: array
           items:
@@ -6218,6 +6222,10 @@ components:
           $ref: '#/components/schemas/OpenAIResponseUsage'
           description: >-
             (Optional) Token usage information for the response
+        instructions:
+          type: string
+          description: >-
+            (Optional) System message inserted into the model's context
       additionalProperties: false
       required:
         - created_at

docs/static/stainless-llama-stack-spec.html

@@ -9272,6 +9272,10 @@
                         "$ref": "#/components/schemas/OpenAIResponseUsage",
                         "description": "(Optional) Token usage information for the response"
                     },
+                    "instructions": {
+                        "type": "string",
+                        "description": "(Optional) System message inserted into the model's context"
+                    },
                     "input": {
                         "type": "array",
                         "items": {
@@ -9820,6 +9824,10 @@
                     "usage": {
                         "$ref": "#/components/schemas/OpenAIResponseUsage",
                         "description": "(Optional) Token usage information for the response"
+                    },
+                    "instructions": {
+                        "type": "string",
+                        "description": "(Optional) System message inserted into the model's context"
                     }
                 },
                 "additionalProperties": false,

docs/static/stainless-llama-stack-spec.yaml

@@ -7028,6 +7028,10 @@ components:
           $ref: '#/components/schemas/OpenAIResponseUsage'
           description: >-
             (Optional) Token usage information for the response
+        instructions:
+          type: string
+          description: >-
+            (Optional) System message inserted into the model's context
         input:
           type: array
           items:
@@ -7431,6 +7435,10 @@ components:
           $ref: '#/components/schemas/OpenAIResponseUsage'
           description: >-
             (Optional) Token usage information for the response
+        instructions:
+          type: string
+          description: >-
+            (Optional) System message inserted into the model's context
       additionalProperties: false
       required:
         - created_at

llama_stack/apis/agents/openai_responses.py

@@ -545,6 +545,7 @@ class OpenAIResponseObject(BaseModel):
     :param tools: (Optional) An array of tools the model may call while generating a response.
     :param truncation: (Optional) Truncation strategy applied to the response
     :param usage: (Optional) Token usage information for the response
+    :param instructions: (Optional) System message inserted into the model's context
     """
 
     created_at: int
@@ -564,6 +565,7 @@ class OpenAIResponseObject(BaseModel):
     tools: list[OpenAIResponseTool] | None = None
     truncation: str | None = None
     usage: OpenAIResponseUsage | None = None
+    instructions: str | None = None
 
 
 @json_schema_type

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

@@ -359,6 +359,7 @@ class OpenAIResponsesImpl:
             tool_executor=self.tool_executor,
             safety_api=self.safety_api,
             guardrail_ids=guardrail_ids,
+            instructions=instructions,
         )
 
         # Stream the response

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

@@ -110,6 +110,7 @@ class StreamingResponseOrchestrator:
         text: OpenAIResponseText,
         max_infer_iters: int,
         tool_executor,  # Will be the tool execution logic from the main class
+        instructions: str,
         safety_api,
         guardrail_ids: list[str] | None = None,
     ):
@@ -133,6 +134,8 @@ class StreamingResponseOrchestrator:
         self.accumulated_usage: OpenAIResponseUsage | None = None
         # Track if we've sent a refusal response
         self.violation_detected = False
+        # system message that is inserted into the model's context
+        self.instructions = instructions
 
     async def _create_refusal_response(self, violation_message: str) -> OpenAIResponseObjectStream:
         """Create a refusal response to replace streaming content."""
@@ -176,6 +179,7 @@ class StreamingResponseOrchestrator:
             tools=self.ctx.available_tools(),
             error=error,
             usage=self.accumulated_usage,
+            instructions=self.instructions,
         )
 
     async def create_response(self) -> AsyncIterator[OpenAIResponseObjectStream]:

tests/integration/agents/recordings/00f8a71ccb939737ed72a289eede62998c6882727519858bbd5954307d10a673.json

@@ -548,5 +548,6 @@
       }
     ],
     "is_streaming": true
-  }
+  },
+  "id_normalization_mapping": {}
 }

tests/integration/agents/recordings/0940d1521204120ff9687b8ad6bf54c271f879db8b5a6ce62848b86a43bc49e4.json

@@ -0,0 +1,447 @@
+{
+  "test_id": "tests/integration/agents/test_agents.py::test_tool_choice_get_boiling_point[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": "system",
+          "content": "You are a helpful assistant"
+        },
+        {
+          "role": "user",
+          "content": "What is the boiling point of the liquid polyjuice in celsius?"
+        },
+        {
+          "role": "assistant",
+          "content": "",
+          "tool_calls": [
+            {
+              "id": "call_5qverjg6",
+              "type": "function",
+              "function": {
+                "name": "get_boiling_point",
+                "arguments": "{\"celcius\":true,\"liquid_name\":\"polyjuice\"}"
+              }
+            }
+          ]
+        },
+        {
+          "role": "tool",
+          "tool_call_id": "call_5qverjg6",
+          "content": "-100"
+        }
+      ],
+      "max_tokens": 512,
+      "stream": true,
+      "temperature": 0.0001,
+      "tool_choice": {
+        "type": "function",
+        "function": {
+          "name": "get_boiling_point"
+        }
+      },
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "get_boiling_point",
+            "description": "Returns the boiling point of a liquid in Celcius or Fahrenheit.",
+            "parameters": {
+              "type": "object",
+              "properties": {
+                "liquid_name": {
+                  "type": "string",
+                  "description": "The name of the liquid"
+                },
+                "celcius": {
+                  "type": "boolean",
+                  "description": "Whether to return the boiling point in Celcius"
+                }
+              },
+              "required": [
+                "liquid_name"
+              ]
+            }
+          }
+        }
+      ],
+      "top_p": 0.9
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "llama3.2:3b-instruct-fp16"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-0940d1521204",
+          "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-0940d1521204",
+          "choices": [
+            {
+              "delta": {
+                "content": " boiling",
+                "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-0940d1521204",
+          "choices": [
+            {
+              "delta": {
+                "content": " point",
+                "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-0940d1521204",
+          "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-0940d1521204",
+          "choices": [
+            {
+              "delta": {
+                "content": " liquid",
+                "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-0940d1521204",
+          "choices": [
+            {
+              "delta": {
+                "content": " poly",
+                "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-0940d1521204",
+          "choices": [
+            {
+              "delta": {
+                "content": "ju",
+                "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-0940d1521204",
+          "choices": [
+            {
+              "delta": {
+                "content": "ice",
+                "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-0940d1521204",
+          "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-0940d1521204",
+          "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-0940d1521204",
+          "choices": [
+            {
+              "delta": {
+                "content": "100",
+                "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-0940d1521204",
+          "choices": [
+            {
+              "delta": {
+                "content": "\u00b0C",
+                "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-0940d1521204",
+          "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-0940d1521204",
+          "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
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/agents/recordings/0f5443c07d1568fd139b8f3ea0aaa3de23d22b30f353c8ed7e6cfd033d904e04.json

@@ -0,0 +1,888 @@
+{
+  "test_id": "tests/integration/agents/test_openai_responses.py::test_response_with_instructions[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": "system",
+          "content": "You are a helpful assistant and speak in pirate language."
+        },
+        {
+          "role": "user",
+          "content": "What is the capital of France?"
+        },
+        {
+          "role": "assistant",
+          "content": "The capital of France is Paris."
+        }
+      ],
+      "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-0f5443c07d15",
+          "choices": [
+            {
+              "delta": {
+                "content": " Yer",
+                "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-0f5443c07d15",
+          "choices": [
+            {
+              "delta": {
+                "content": " look",
+                "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-0f5443c07d15",
+          "choices": [
+            {
+              "delta": {
+                "content": "in",
+                "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-0f5443c07d15",
+          "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-0f5443c07d15",
+          "choices": [
+            {
+              "delta": {
+                "content": " fer",
+                "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-0f5443c07d15",
+          "choices": [
+            {
+              "delta": {
+                "content": " a",
+                "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-0f5443c07d15",
+          "choices": [
+            {
+              "delta": {
+                "content": " port",
+                "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-0f5443c07d15",
+          "choices": [
+            {
+              "delta": {
+                "content": " o",
+                "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-0f5443c07d15",
+          "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-0f5443c07d15",
+          "choices": [
+            {
+              "delta": {
+                "content": " call",
+                "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-0f5443c07d15",
+          "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-0f5443c07d15",
+          "choices": [
+            {
+              "delta": {
+                "content": " eh",
+                "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-0f5443c07d15",
+          "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-0f5443c07d15",
+          "choices": [
+            {
+              "delta": {
+                "content": " That",
+                "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-0f5443c07d15",
+          "choices": [
+            {
+              "delta": {
+                "content": " be",
+                "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-0f5443c07d15",
+          "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-0f5443c07d15",
+          "choices": [
+            {
+              "delta": {
+                "content": " one",
+                "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-0f5443c07d15",
+          "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-0f5443c07d15",
+          "choices": [
+            {
+              "delta": {
+                "content": " Yer",
+                "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-0f5443c07d15",
+          "choices": [
+            {
+              "delta": {
+                "content": " won",
+                "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-0f5443c07d15",
+          "choices": [
+            {
+              "delta": {
+                "content": "'t",
+                "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-0f5443c07d15",
+          "choices": [
+            {
+              "delta": {
+                "content": " go",
+                "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-0f5443c07d15",
+          "choices": [
+            {
+              "delta": {
+                "content": " astr",
+                "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-0f5443c07d15",
+          "choices": [
+            {
+              "delta": {
+                "content": "ay",
+                "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-0f5443c07d15",
+          "choices": [
+            {
+              "delta": {
+                "content": " with",
+                "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-0f5443c07d15",
+          "choices": [
+            {
+              "delta": {
+                "content": " that",
+                "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-0f5443c07d15",
+          "choices": [
+            {
+              "delta": {
+                "content": " answer",
+                "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-0f5443c07d15",
+          "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-0f5443c07d15",
+          "choices": [
+            {
+              "delta": {
+                "content": " mate",
+                "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-0f5443c07d15",
+          "choices": [
+            {
+              "delta": {
+                "content": "y",
+                "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-0f5443c07d15",
+          "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-0f5443c07d15",
+          "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-0f5443c07d15",
+          "choices": [],
+          "created": 0,
+          "model": "llama3.2:3b-instruct-fp16",
+          "object": "chat.completion.chunk",
+          "service_tier": null,
+          "system_fingerprint": "fp_ollama",
+          "usage": {
+            "completion_tokens": 32,
+            "prompt_tokens": 50,
+            "total_tokens": 82,
+            "completion_tokens_details": null,
+            "prompt_tokens_details": null
+          }
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/agents/recordings/13fac3724cd626a119153f60fa56bb54955fe0b10f5c4102b78e2d428b5ffc7a.json

@@ -54,5 +54,6 @@
       }
     },
     "is_streaming": false
-  }
+  },
+  "id_normalization_mapping": {}
 }

tests/integration/agents/recordings/15b23045b5cdfc49228d58e4a082f8402f7c91d15e2240f855cc9b8b4e25352a.json

@@ -0,0 +1,256 @@
+{
+  "test_id": "tests/integration/agents/test_openai_responses.py::test_response_with_instructions[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-15b23045b5cd",
+          "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-15b23045b5cd",
+          "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-15b23045b5cd",
+          "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-15b23045b5cd",
+          "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-15b23045b5cd",
+          "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-15b23045b5cd",
+          "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-15b23045b5cd",
+          "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-15b23045b5cd",
+          "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-15b23045b5cd",
+          "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/1a0d3109cf92111ed4cb061a857dee0b99fa1e0b27934de1e6c5d29c03026626.json

@@ -54,5 +54,6 @@
       }
     },
     "is_streaming": false
-  }
+  },
+  "id_normalization_mapping": {}
 }

tests/integration/agents/recordings/1f0aef7475448c77021b4e321125b4df3aadc1637a93358a85c5ec2de8338332.json

@@ -0,0 +1,442 @@
+{
+  "test_id": "tests/integration/agents/test_agents.py::test_create_turn_response[ollama/llama3.2:3b-instruct-fp16-client_tools1]",
+  "request": {
+    "method": "POST",
+    "url": "http://0.0.0.0:11434/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "llama3.2:3b-instruct-fp16",
+      "messages": [
+        {
+          "role": "system",
+          "content": "You are a helpful assistant"
+        },
+        {
+          "role": "user",
+          "content": "Call get_boiling_point_with_metadata tool and answer What is the boiling point of polyjuice?"
+        },
+        {
+          "role": "assistant",
+          "content": "",
+          "tool_calls": [
+            {
+              "id": "call_klhbln13",
+              "type": "function",
+              "function": {
+                "name": "get_boiling_point_with_metadata",
+                "arguments": "{\"celcius\":false,\"liquid_name\":\"polyjuice\"}"
+              }
+            }
+          ]
+        },
+        {
+          "role": "tool",
+          "tool_call_id": "call_klhbln13",
+          "content": "-212"
+        }
+      ],
+      "max_tokens": 512,
+      "stream": true,
+      "temperature": 0.0001,
+      "tool_choice": "auto",
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "get_boiling_point_with_metadata",
+            "description": "Returns the boiling point of a liquid in Celcius or Fahrenheit",
+            "parameters": {
+              "type": "object",
+              "properties": {
+                "liquid_name": {
+                  "type": "string",
+                  "description": "The name of the liquid"
+                },
+                "celcius": {
+                  "type": "boolean",
+                  "description": "Whether to return the boiling point in Celcius"
+                }
+              },
+              "required": [
+                "liquid_name"
+              ]
+            }
+          }
+        }
+      ],
+      "top_p": 0.9
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "llama3.2:3b-instruct-fp16"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-1f0aef747544",
+          "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-1f0aef747544",
+          "choices": [
+            {
+              "delta": {
+                "content": " boiling",
+                "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-1f0aef747544",
+          "choices": [
+            {
+              "delta": {
+                "content": " point",
+                "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-1f0aef747544",
+          "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-1f0aef747544",
+          "choices": [
+            {
+              "delta": {
+                "content": " poly",
+                "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-1f0aef747544",
+          "choices": [
+            {
+              "delta": {
+                "content": "ju",
+                "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-1f0aef747544",
+          "choices": [
+            {
+              "delta": {
+                "content": "ice",
+                "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-1f0aef747544",
+          "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-1f0aef747544",
+          "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-1f0aef747544",
+          "choices": [
+            {
+              "delta": {
+                "content": "212",
+                "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-1f0aef747544",
+          "choices": [
+            {
+              "delta": {
+                "content": " degrees",
+                "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-1f0aef747544",
+          "choices": [
+            {
+              "delta": {
+                "content": " Celsius",
+                "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-1f0aef747544",
+          "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-1f0aef747544",
+          "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
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

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

@@ -56,7 +56,7 @@
                 "tool_calls": [
                   {
                     "index": 0,
-                    "id": "call_os3xa9go",
+                    "id": "call_6nqo069h",
                     "function": {
                       "arguments": "{\"city\":\"Tokyo\"}",
                       "name": "get_weather"
@@ -115,9 +115,9 @@
           "service_tier": null,
           "system_fingerprint": "fp_ollama",
           "usage": {
-            "completion_tokens": 15,
+            "completion_tokens": 18,
             "prompt_tokens": 179,
-            "total_tokens": 194,
+            "total_tokens": 197,
             "completion_tokens_details": null,
             "prompt_tokens_details": null
           }

tests/integration/agents/recordings/256d8571909664fc6c925058b2ff1b1d0e0bd618975cbf4752eb31ada6d2482b.json

@@ -0,0 +1,416 @@
+{
+  "test_id": "tests/integration/agents/test_agents.py::test_custom_tool_infinite_loop[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": "system",
+          "content": "You are a helpful assistant Always respond with tool calls no matter what. "
+        },
+        {
+          "role": "user",
+          "content": "Get the boiling point of polyjuice with a tool call."
+        },
+        {
+          "role": "assistant",
+          "content": "",
+          "tool_calls": [
+            {
+              "id": "call_9x4z21g1",
+              "type": "function",
+              "function": {
+                "name": "get_boiling_point",
+                "arguments": "{\"celcius\":\"true\",\"liquid_name\":\"polyjuice\"}"
+              }
+            }
+          ]
+        },
+        {
+          "role": "tool",
+          "tool_call_id": "call_9x4z21g1",
+          "content": "-100"
+        }
+      ],
+      "max_tokens": 512,
+      "stream": true,
+      "temperature": 0.0001,
+      "tool_choice": "auto",
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "get_boiling_point",
+            "description": "Returns the boiling point of a liquid in Celcius or Fahrenheit.",
+            "parameters": {
+              "type": "object",
+              "properties": {
+                "liquid_name": {
+                  "type": "string",
+                  "description": "The name of the liquid"
+                },
+                "celcius": {
+                  "type": "boolean",
+                  "description": "Whether to return the boiling point in Celcius"
+                }
+              },
+              "required": [
+                "liquid_name"
+              ]
+            }
+          }
+        }
+      ],
+      "top_p": 0.9
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "llama3.2:3b-instruct-fp16"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-256d85719096",
+          "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-256d85719096",
+          "choices": [
+            {
+              "delta": {
+                "content": " boiling",
+                "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-256d85719096",
+          "choices": [
+            {
+              "delta": {
+                "content": " point",
+                "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-256d85719096",
+          "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-256d85719096",
+          "choices": [
+            {
+              "delta": {
+                "content": " Poly",
+                "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-256d85719096",
+          "choices": [
+            {
+              "delta": {
+                "content": "ju",
+                "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-256d85719096",
+          "choices": [
+            {
+              "delta": {
+                "content": "ice",
+                "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-256d85719096",
+          "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-256d85719096",
+          "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-256d85719096",
+          "choices": [
+            {
+              "delta": {
+                "content": "100",
+                "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-256d85719096",
+          "choices": [
+            {
+              "delta": {
+                "content": "\u00b0C",
+                "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-256d85719096",
+          "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-256d85719096",
+          "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
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/agents/recordings/292308724331c7172aaf91fe1373f2fbc9d626af08314bd7f5ba69d038ea7c1b.json

@@ -54,5 +54,6 @@
       }
     },
     "is_streaming": false
-  }
+  },
+  "id_normalization_mapping": {}
 }

tests/integration/agents/recordings/36e22908b34c0835037ba7b52477c5db69585e66f4fde18eaa8bfd4bb4e3d783.json

@@ -66,7 +66,7 @@
                 "tool_calls": [
                   {
                     "index": 0,
-                    "id": "call_ixvkq8fh",
+                    "id": "call_icfpgg5q",
                     "function": {
                       "arguments": "{\"celcius\":true,\"liquid_name\":\"polyjuice\"}",
                       "name": "get_boiling_point"
@@ -116,5 +116,6 @@
       }
     ],
     "is_streaming": true
-  }
+  },
+  "id_normalization_mapping": {}
 }

tests/integration/agents/recordings/45175e711385e24f62516c3982eaf8fb6bbab4f37691fadc08812ce223dfc628.json

@@ -54,5 +54,6 @@
       }
     },
     "is_streaming": false
-  }
+  },
+  "id_normalization_mapping": {}
 }

tests/integration/agents/recordings/45d0aabc502385b4cc23e16706a1c594644b2a1387ac3cee7cd434df25e8f22f.json

@@ -0,0 +1,442 @@
+{
+  "test_id": "tests/integration/agents/test_agents.py::test_custom_tool[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": "system",
+          "content": "You are a helpful assistant"
+        },
+        {
+          "role": "user",
+          "content": "What is the boiling point of the liquid polyjuice in celsius?"
+        },
+        {
+          "role": "assistant",
+          "content": "",
+          "tool_calls": [
+            {
+              "id": "call_icfpgg5q",
+              "type": "function",
+              "function": {
+                "name": "get_boiling_point",
+                "arguments": "{\"celcius\":true,\"liquid_name\":\"polyjuice\"}"
+              }
+            }
+          ]
+        },
+        {
+          "role": "tool",
+          "tool_call_id": "call_icfpgg5q",
+          "content": "-100"
+        }
+      ],
+      "max_tokens": 512,
+      "stream": true,
+      "temperature": 0.0001,
+      "tool_choice": "auto",
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "get_boiling_point",
+            "description": "Returns the boiling point of a liquid in Celcius or Fahrenheit.",
+            "parameters": {
+              "type": "object",
+              "properties": {
+                "liquid_name": {
+                  "type": "string",
+                  "description": "The name of the liquid"
+                },
+                "celcius": {
+                  "type": "boolean",
+                  "description": "Whether to return the boiling point in Celcius"
+                }
+              },
+              "required": [
+                "liquid_name"
+              ]
+            }
+          }
+        }
+      ],
+      "top_p": 0.9
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "llama3.2:3b-instruct-fp16"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-45d0aabc5023",
+          "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-45d0aabc5023",
+          "choices": [
+            {
+              "delta": {
+                "content": " boiling",
+                "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-45d0aabc5023",
+          "choices": [
+            {
+              "delta": {
+                "content": " point",
+                "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-45d0aabc5023",
+          "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-45d0aabc5023",
+          "choices": [
+            {
+              "delta": {
+                "content": " liquid",
+                "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-45d0aabc5023",
+          "choices": [
+            {
+              "delta": {
+                "content": " poly",
+                "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-45d0aabc5023",
+          "choices": [
+            {
+              "delta": {
+                "content": "ju",
+                "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-45d0aabc5023",
+          "choices": [
+            {
+              "delta": {
+                "content": "ice",
+                "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-45d0aabc5023",
+          "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-45d0aabc5023",
+          "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-45d0aabc5023",
+          "choices": [
+            {
+              "delta": {
+                "content": "100",
+                "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-45d0aabc5023",
+          "choices": [
+            {
+              "delta": {
+                "content": "\u00b0C",
+                "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-45d0aabc5023",
+          "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-45d0aabc5023",
+          "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
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

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

@@ -48,7 +48,7 @@
                 "tool_calls": [
                   {
                     "index": 0,
-                    "id": "call_lqrdy0rt",
+                    "id": "call_x427af31",
                     "function": {
                       "arguments": "{}",
                       "name": "get_current_time"
@@ -107,9 +107,9 @@
           "service_tier": null,
           "system_fingerprint": "fp_ollama",
           "usage": {
-            "completion_tokens": 14,
+            "completion_tokens": 12,
             "prompt_tokens": 161,
-            "total_tokens": 175,
+            "total_tokens": 173,
             "completion_tokens_details": null,
             "prompt_tokens_details": null
           }

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

@@ -56,7 +56,7 @@
                 "tool_calls": [
                   {
                     "index": 0,
-                    "id": "call_4ibtjudr",
+                    "id": "call_wkjhgmpf",
                     "function": {
                       "arguments": "{\"city\":\"Tokyo\"}",
                       "name": "get_weather"

tests/integration/agents/recordings/697a25dd7f0ff515f567c883ad72ae9dca423726834aec8b38420dccb735c050.json

@@ -66,7 +66,7 @@
                 "tool_calls": [
                   {
                     "index": 0,
-                    "id": "call_pojpzwm8",
+                    "id": "call_klhbln13",
                     "function": {
                       "arguments": "{\"celcius\":false,\"liquid_name\":\"polyjuice\"}",
                       "name": "get_boiling_point_with_metadata"
@@ -116,5 +116,6 @@
       }
     ],
     "is_streaming": true
-  }
+  },
+  "id_normalization_mapping": {}
 }

tests/integration/agents/recordings/6da760645fe224ace4ab628e4f647259897598e28037fe5f7c09f6677edd08e9.json

@@ -54,5 +54,6 @@
       }
     },
     "is_streaming": false
-  }
+  },
+  "id_normalization_mapping": {}
 }

tests/integration/agents/recordings/7094319e038424fbec54338c397b487c7128fc28534351deb4662fba31043fa4.json

@@ -54,5 +54,6 @@
       }
     },
     "is_streaming": false
-  }
+  },
+  "id_normalization_mapping": {}
 }

tests/integration/agents/recordings/74c26f63592ceedb76eb9623fce41773965dd66b569506b5622b1a797c45f8e4.json

@@ -54,5 +54,6 @@
       }
     },
     "is_streaming": false
-  }
+  },
+  "id_normalization_mapping": {}
 }

tests/integration/agents/recordings/7e0d8c4abe407611ceaa99beea2e9973e2a938cab3db3e1be017bbe8be10edc6.json

@@ -66,7 +66,7 @@
                 "tool_calls": [
                   {
                     "index": 0,
-                    "id": "call_rwasjr3y",
+                    "id": "call_zqu5i0ti",
                     "function": {
                       "arguments": "{\"celcius\":null,\"liquid_name\":\"polyjuice\"}",
                       "name": "get_boiling_point"
@@ -116,5 +116,6 @@
       }
     ],
     "is_streaming": true
-  }
+  },
+  "id_normalization_mapping": {}
 }

tests/integration/agents/recordings/7e794c73bf79604a10482bba03124849cb763c7bb66acf3937b524a539b80366.json

@@ -0,0 +1,59 @@
+{
+  "test_id": "tests/integration/agents/test_agents.py::test_tool_choice_required[ollama/llama3.2:3b-instruct-fp16]",
+  "request": {
+    "method": "POST",
+    "url": "http://0.0.0.0:11434/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "llama-guard3:1b",
+      "messages": [
+        {
+          "role": "user",
+          "content": "Task: Check if there is unsafe content in 'Assistant' messages in conversations according our safety policy with the below categories.\n\n<BEGIN UNSAFE CONTENT CATEGORIES>\nS1: Violent Crimes.\nS2: Non-Violent Crimes.\nS3: Sex Crimes.\nS4: Child Exploitation.\nS5: Defamation.\nS6: Specialized Advice.\nS7: Privacy.\nS8: Intellectual Property.\nS9: Indiscriminate Weapons.\nS10: Hate.\nS11: Self-Harm.\nS12: Sexual Content.\nS13: Elections.\nS14: Code Interpreter Abuse.\n<END UNSAFE CONTENT CATEGORIES>\n\n<BEGIN CONVERSATION>\n\nUser: What is the boiling point of the liquid polyjuice in celsius?\n\nAssistant: \n\nTool: -100\n\nAssistant: The boiling point of liquid polyjuice is -100\u00b0C.\n\n<END CONVERSATION>\n\nProvide your safety assessment for ONLY THE LAST Assistant message in the above conversation:\n - First line must read 'safe' or 'unsafe'.\n - If unsafe, a second line must include a comma-separated list of violated categories."
+        }
+      ],
+      "stream": false,
+      "temperature": 0.0
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "llama-guard3:1b"
+  },
+  "response": {
+    "body": {
+      "__type__": "openai.types.chat.chat_completion.ChatCompletion",
+      "__data__": {
+        "id": "rec-7e794c73bf79",
+        "choices": [
+          {
+            "finish_reason": "stop",
+            "index": 0,
+            "logprobs": null,
+            "message": {
+              "content": "safe",
+              "refusal": null,
+              "role": "assistant",
+              "annotations": null,
+              "audio": null,
+              "function_call": null,
+              "tool_calls": null
+            }
+          }
+        ],
+        "created": 0,
+        "model": "llama-guard3:1b",
+        "object": "chat.completion",
+        "service_tier": null,
+        "system_fingerprint": "fp_ollama",
+        "usage": {
+          "completion_tokens": 2,
+          "prompt_tokens": 421,
+          "total_tokens": 423,
+          "completion_tokens_details": null,
+          "prompt_tokens_details": null
+        }
+      }
+    },
+    "is_streaming": false
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/agents/recordings/868820c8d798c0d16063d1750a65ae2632ecf543ee440e7d87ea16f8e83461a5.json

@@ -54,5 +54,6 @@
       }
     },
     "is_streaming": false
-  }
+  },
+  "id_normalization_mapping": {}
 }

tests/integration/agents/recordings/86e2b939aabb9dfe7ec712a6b20a5809d6fb56f8c9f92d93030f57cba51a1fe2.json

@@ -54,5 +54,6 @@
       }
     },
     "is_streaming": false
-  }
+  },
+  "id_normalization_mapping": {}
 }

tests/integration/agents/recordings/8c4ec47152697a5b34e44d75af581efbada34e06f59ddf536149871c64c9a247.json

@@ -0,0 +1,442 @@
+{
+  "test_id": "tests/integration/agents/test_agents.py::test_create_turn_response[ollama/llama3.2:3b-instruct-fp16-client_tools0]",
+  "request": {
+    "method": "POST",
+    "url": "http://0.0.0.0:11434/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "llama3.2:3b-instruct-fp16",
+      "messages": [
+        {
+          "role": "system",
+          "content": "You are a helpful assistant"
+        },
+        {
+          "role": "user",
+          "content": "Call get_boiling_point tool and answer What is the boiling point of polyjuice?"
+        },
+        {
+          "role": "assistant",
+          "content": "",
+          "tool_calls": [
+            {
+              "id": "call_zqu5i0ti",
+              "type": "function",
+              "function": {
+                "name": "get_boiling_point",
+                "arguments": "{\"celcius\":null,\"liquid_name\":\"polyjuice\"}"
+              }
+            }
+          ]
+        },
+        {
+          "role": "tool",
+          "tool_call_id": "call_zqu5i0ti",
+          "content": "-212"
+        }
+      ],
+      "max_tokens": 512,
+      "stream": true,
+      "temperature": 0.0001,
+      "tool_choice": "auto",
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "get_boiling_point",
+            "description": "Returns the boiling point of a liquid in Celcius or Fahrenheit.",
+            "parameters": {
+              "type": "object",
+              "properties": {
+                "liquid_name": {
+                  "type": "string",
+                  "description": "The name of the liquid"
+                },
+                "celcius": {
+                  "type": "boolean",
+                  "description": "Whether to return the boiling point in Celcius"
+                }
+              },
+              "required": [
+                "liquid_name"
+              ]
+            }
+          }
+        }
+      ],
+      "top_p": 0.9
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "llama3.2:3b-instruct-fp16"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-8c4ec4715269",
+          "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-8c4ec4715269",
+          "choices": [
+            {
+              "delta": {
+                "content": " boiling",
+                "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-8c4ec4715269",
+          "choices": [
+            {
+              "delta": {
+                "content": " point",
+                "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-8c4ec4715269",
+          "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-8c4ec4715269",
+          "choices": [
+            {
+              "delta": {
+                "content": " poly",
+                "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-8c4ec4715269",
+          "choices": [
+            {
+              "delta": {
+                "content": "ju",
+                "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-8c4ec4715269",
+          "choices": [
+            {
+              "delta": {
+                "content": "ice",
+                "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-8c4ec4715269",
+          "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-8c4ec4715269",
+          "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-8c4ec4715269",
+          "choices": [
+            {
+              "delta": {
+                "content": "212",
+                "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-8c4ec4715269",
+          "choices": [
+            {
+              "delta": {
+                "content": " degrees",
+                "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-8c4ec4715269",
+          "choices": [
+            {
+              "delta": {
+                "content": " Celsius",
+                "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-8c4ec4715269",
+          "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-8c4ec4715269",
+          "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
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/agents/recordings/8ed0947593196c2b2f68035e248c137813e8db50d0d46395ef9ba98636fa5819.json

@@ -54,5 +54,6 @@
       }
     },
     "is_streaming": false
-  }
+  },
+  "id_normalization_mapping": {}
 }

tests/integration/agents/recordings/8fc418c02b8b6fe09238e36fb72c5e7fc9d41fdaa3eb357f518e0fcaec5ea1e2.json

@@ -0,0 +1,59 @@
+{
+  "test_id": "tests/integration/agents/test_agents.py::test_tool_choice_get_boiling_point[ollama/llama3.2:3b-instruct-fp16]",
+  "request": {
+    "method": "POST",
+    "url": "http://0.0.0.0:11434/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "llama-guard3:1b",
+      "messages": [
+        {
+          "role": "user",
+          "content": "Task: Check if there is unsafe content in 'Assistant' messages in conversations according our safety policy with the below categories.\n\n<BEGIN UNSAFE CONTENT CATEGORIES>\nS1: Violent Crimes.\nS2: Non-Violent Crimes.\nS3: Sex Crimes.\nS4: Child Exploitation.\nS5: Defamation.\nS6: Specialized Advice.\nS7: Privacy.\nS8: Intellectual Property.\nS9: Indiscriminate Weapons.\nS10: Hate.\nS11: Self-Harm.\nS12: Sexual Content.\nS13: Elections.\nS14: Code Interpreter Abuse.\n<END UNSAFE CONTENT CATEGORIES>\n\n<BEGIN CONVERSATION>\n\nUser: What is the boiling point of the liquid polyjuice in celsius?\n\nAssistant: \n\nTool: -100\n\nAssistant: The boiling point of liquid polyjuice is -100\u00b0C.\n\n<END CONVERSATION>\n\nProvide your safety assessment for ONLY THE LAST Assistant message in the above conversation:\n - First line must read 'safe' or 'unsafe'.\n - If unsafe, a second line must include a comma-separated list of violated categories."
+        }
+      ],
+      "stream": false,
+      "temperature": 0.0
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "llama-guard3:1b"
+  },
+  "response": {
+    "body": {
+      "__type__": "openai.types.chat.chat_completion.ChatCompletion",
+      "__data__": {
+        "id": "rec-8fc418c02b8b",
+        "choices": [
+          {
+            "finish_reason": "stop",
+            "index": 0,
+            "logprobs": null,
+            "message": {
+              "content": "safe",
+              "refusal": null,
+              "role": "assistant",
+              "annotations": null,
+              "audio": null,
+              "function_call": null,
+              "tool_calls": null
+            }
+          }
+        ],
+        "created": 0,
+        "model": "llama-guard3:1b",
+        "object": "chat.completion",
+        "service_tier": null,
+        "system_fingerprint": "fp_ollama",
+        "usage": {
+          "completion_tokens": 2,
+          "prompt_tokens": 421,
+          "total_tokens": 423,
+          "completion_tokens_details": null,
+          "prompt_tokens_details": null
+        }
+      }
+    },
+    "is_streaming": false
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/agents/recordings/901956b3a51b792f2506d603489af51636b480db9cc520614ee4886418776237.json

@@ -54,5 +54,6 @@
       }
     },
     "is_streaming": false
-  }
+  },
+  "id_normalization_mapping": {}
 }

tests/integration/agents/recordings/958f9b74e98bcf41e4988db8ad15494b8fe0ff707261108305353e4ad980195f.json

@@ -1510,5 +1510,6 @@
       }
     ],
     "is_streaming": true
-  }
+  },
+  "id_normalization_mapping": {}
 }

tests/integration/agents/recordings/96623a251d6e51ee6ba21c53ca111d4aa54882a124d783a8096fd88adf481065.json

@@ -54,5 +54,6 @@
       }
     },
     "is_streaming": false
-  }
+  },
+  "id_normalization_mapping": {}
 }

tests/integration/agents/recordings/a702e4bf918e94acd0d76ed753c120a4704dde82acf5ae198982fd62bd103279.json

@@ -54,5 +54,6 @@
       }
     },
     "is_streaming": false
-  }
+  },
+  "id_normalization_mapping": {}
 }

tests/integration/agents/recordings/ad3f6a2b4031bcd38026c3c50617851f102c12946164a563584e6316bd1b6228.json

@@ -54,5 +54,6 @@
       }
     },
     "is_streaming": false
-  }
+  },
+  "id_normalization_mapping": {}
 }

tests/integration/agents/recordings/b3c24a0ab429fb3d7e3680a2a689a8eddb2c2aaf826b513fec55dcd70cdf35ea.json

@@ -0,0 +1,260 @@
+{
+  "test_id": "tests/integration/agents/test_openai_responses.py::test_response_with_instructions[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": "system",
+          "content": "You are a helpful assistant."
+        },
+        {
+          "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-b3c24a0ab429",
+          "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-b3c24a0ab429",
+          "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-b3c24a0ab429",
+          "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-b3c24a0ab429",
+          "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-b3c24a0ab429",
+          "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-b3c24a0ab429",
+          "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-b3c24a0ab429",
+          "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-b3c24a0ab429",
+          "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-b3c24a0ab429",
+          "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": 38,
+            "total_tokens": 46,
+            "completion_tokens_details": null,
+            "prompt_tokens_details": null
+          }
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/agents/recordings/b4a47451a2af579b9dfb4a60bacaee0f274fc53b263c25fe5e9e4bc23739f3db.json

@@ -0,0 +1,442 @@
+{
+  "test_id": "tests/integration/agents/test_agents.py::test_tool_choice_required[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": "system",
+          "content": "You are a helpful assistant"
+        },
+        {
+          "role": "user",
+          "content": "What is the boiling point of the liquid polyjuice in celsius?"
+        },
+        {
+          "role": "assistant",
+          "content": "",
+          "tool_calls": [
+            {
+              "id": "call_z1rt0qb1",
+              "type": "function",
+              "function": {
+                "name": "get_boiling_point",
+                "arguments": "{\"celcius\":true,\"liquid_name\":\"polyjuice\"}"
+              }
+            }
+          ]
+        },
+        {
+          "role": "tool",
+          "tool_call_id": "call_z1rt0qb1",
+          "content": "-100"
+        }
+      ],
+      "max_tokens": 512,
+      "stream": true,
+      "temperature": 0.0001,
+      "tool_choice": "required",
+      "tools": [
+        {
+          "type": "function",
+          "function": {
+            "name": "get_boiling_point",
+            "description": "Returns the boiling point of a liquid in Celcius or Fahrenheit.",
+            "parameters": {
+              "type": "object",
+              "properties": {
+                "liquid_name": {
+                  "type": "string",
+                  "description": "The name of the liquid"
+                },
+                "celcius": {
+                  "type": "boolean",
+                  "description": "Whether to return the boiling point in Celcius"
+                }
+              },
+              "required": [
+                "liquid_name"
+              ]
+            }
+          }
+        }
+      ],
+      "top_p": 0.9
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "llama3.2:3b-instruct-fp16"
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.chat.chat_completion_chunk.ChatCompletionChunk",
+        "__data__": {
+          "id": "rec-b4a47451a2af",
+          "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-b4a47451a2af",
+          "choices": [
+            {
+              "delta": {
+                "content": " boiling",
+                "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-b4a47451a2af",
+          "choices": [
+            {
+              "delta": {
+                "content": " point",
+                "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-b4a47451a2af",
+          "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-b4a47451a2af",
+          "choices": [
+            {
+              "delta": {
+                "content": " liquid",
+                "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-b4a47451a2af",
+          "choices": [
+            {
+              "delta": {
+                "content": " poly",
+                "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-b4a47451a2af",
+          "choices": [
+            {
+              "delta": {
+                "content": "ju",
+                "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-b4a47451a2af",
+          "choices": [
+            {
+              "delta": {
+                "content": "ice",
+                "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-b4a47451a2af",
+          "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-b4a47451a2af",
+          "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-b4a47451a2af",
+          "choices": [
+            {
+              "delta": {
+                "content": "100",
+                "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-b4a47451a2af",
+          "choices": [
+            {
+              "delta": {
+                "content": "\u00b0C",
+                "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-b4a47451a2af",
+          "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-b4a47451a2af",
+          "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
+        }
+      }
+    ],
+    "is_streaming": true
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/agents/recordings/bcb50763cac6527944653e77df1f2cc9e9481ffae9cff5693b2f669270c9c0a7.json

@@ -66,7 +66,7 @@
                 "tool_calls": [
                   {
                     "index": 0,
-                    "id": "call_qryqpevz",
+                    "id": "call_9x4z21g1",
                     "function": {
                       "arguments": "{\"celcius\":\"true\",\"liquid_name\":\"polyjuice\"}",
                       "name": "get_boiling_point"
@@ -116,5 +116,6 @@
       }
     ],
     "is_streaming": true
-  }
+  },
+  "id_normalization_mapping": {}
 }

tests/integration/agents/recordings/c97c102959ea8c64a43e4c752bf5e979c7a298fdbeedec153954ce817da7e3e7.json

@@ -54,5 +54,6 @@
       }
     },
     "is_streaming": false
-  }
+  },
+  "id_normalization_mapping": {}
 }

tests/integration/agents/recordings/ca95f47c2896a7ce1536a3cf1a78170e073e49e478658a4b098b3581a703e843.json

@@ -54,5 +54,6 @@
       }
     },
     "is_streaming": false
-  }
+  },
+  "id_normalization_mapping": {}
 }

tests/integration/agents/recordings/d35fc2ef48595f5d0afe8fb617c8df864a155017e8d4c5d2e2b2c51e9cfaed5e.json

@@ -54,5 +54,6 @@
       }
     },
     "is_streaming": false
-  }
+  },
+  "id_normalization_mapping": {}
 }

tests/integration/agents/recordings/da6fc54bb65dd1f83e577109b8228b644e76999ebb2fe8f09bead3dee56a6046.json

@@ -0,0 +1,59 @@
+{
+  "test_id": "tests/integration/agents/test_agents.py::test_custom_tool[ollama/llama3.2:3b-instruct-fp16]",
+  "request": {
+    "method": "POST",
+    "url": "http://0.0.0.0:11434/v1/v1/chat/completions",
+    "headers": {},
+    "body": {
+      "model": "llama-guard3:1b",
+      "messages": [
+        {
+          "role": "user",
+          "content": "Task: Check if there is unsafe content in 'Assistant' messages in conversations according our safety policy with the below categories.\n\n<BEGIN UNSAFE CONTENT CATEGORIES>\nS1: Violent Crimes.\nS2: Non-Violent Crimes.\nS3: Sex Crimes.\nS4: Child Exploitation.\nS5: Defamation.\nS6: Specialized Advice.\nS7: Privacy.\nS8: Intellectual Property.\nS9: Indiscriminate Weapons.\nS10: Hate.\nS11: Self-Harm.\nS12: Sexual Content.\nS13: Elections.\nS14: Code Interpreter Abuse.\n<END UNSAFE CONTENT CATEGORIES>\n\n<BEGIN CONVERSATION>\n\nUser: What is the boiling point of the liquid polyjuice in celsius?\n\nAssistant: \n\nTool: -100\n\nAssistant: The boiling point of liquid polyjuice is -100\u00b0C.\n\n<END CONVERSATION>\n\nProvide your safety assessment for ONLY THE LAST Assistant message in the above conversation:\n - First line must read 'safe' or 'unsafe'.\n - If unsafe, a second line must include a comma-separated list of violated categories."
+        }
+      ],
+      "stream": false,
+      "temperature": 0.0
+    },
+    "endpoint": "/v1/chat/completions",
+    "model": "llama-guard3:1b"
+  },
+  "response": {
+    "body": {
+      "__type__": "openai.types.chat.chat_completion.ChatCompletion",
+      "__data__": {
+        "id": "rec-da6fc54bb65d",
+        "choices": [
+          {
+            "finish_reason": "stop",
+            "index": 0,
+            "logprobs": null,
+            "message": {
+              "content": "safe",
+              "refusal": null,
+              "role": "assistant",
+              "annotations": null,
+              "audio": null,
+              "function_call": null,
+              "tool_calls": null
+            }
+          }
+        ],
+        "created": 0,
+        "model": "llama-guard3:1b",
+        "object": "chat.completion",
+        "service_tier": null,
+        "system_fingerprint": "fp_ollama",
+        "usage": {
+          "completion_tokens": 2,
+          "prompt_tokens": 421,
+          "total_tokens": 423,
+          "completion_tokens_details": null,
+          "prompt_tokens_details": null
+        }
+      }
+    },
+    "is_streaming": false
+  },
+  "id_normalization_mapping": {}
+}

tests/integration/agents/recordings/db5c89b87eba0d129ad9ed17306d4016aeeaf2bbeeaa5643d9620f5ea484430e.json

@@ -71,7 +71,7 @@
                 "tool_calls": [
                   {
                     "index": 0,
-                    "id": "call_ur5tbdbt",
+                    "id": "call_5qverjg6",
                     "function": {
                       "arguments": "{\"celcius\":true,\"liquid_name\":\"polyjuice\"}",
                       "name": "get_boiling_point"
@@ -121,5 +121,6 @@
       }
     ],
     "is_streaming": true
-  }
+  },
+  "id_normalization_mapping": {}
 }

tests/integration/agents/recordings/ed76dd5fdf892c9cc959b2d301a256f81c43a906a0a56684ca97e848f8d6a94c.json

@@ -54,5 +54,6 @@
       }
     },
     "is_streaming": false
-  }
+  },
+  "id_normalization_mapping": {}
 }

tests/integration/agents/recordings/f85c3c14185386eecd4939eeb6b3a3cee734d69beb7cd6d13a3d3c2c64eca734.json

@@ -66,7 +66,7 @@
                 "tool_calls": [
                   {
                     "index": 0,
-                    "id": "call_rq1pcgq7",
+                    "id": "call_z1rt0qb1",
                     "function": {
                       "arguments": "{\"celcius\":true,\"liquid_name\":\"polyjuice\"}",
                       "name": "get_boiling_point"
@@ -116,5 +116,6 @@
       }
     ],
     "is_streaming": true
-  }
+  },
+  "id_normalization_mapping": {}
 }

tests/integration/agents/test_openai_responses.py

@@ -466,3 +466,53 @@ def test_guardrails_with_tools(compat_client, text_model_id):
     # Response should be either a function call or a message
     output_type = response.output[0].type
     assert output_type in ["function_call", "message"]
+
+
+def test_response_with_instructions(openai_client, client_with_models, text_model_id):
+    """Test instructions parameter in the responses object."""
+    if isinstance(client_with_models, LlamaStackAsLibraryClient):
+        pytest.skip("OpenAI responses are not supported when testing with library client yet.")
+
+    client = openai_client
+
+    messages = [
+        {
+            "role": "user",
+            "content": "What is the capital of France?",
+        }
+    ]
+
+    # First create a response without instructions parameter
+    response_w_o_instructions = client.responses.create(
+        model=text_model_id,
+        input=messages,
+        stream=False,
+    )
+
+    # Verify we have None in the instructions field
+    assert response_w_o_instructions.instructions is None
+
+    # Next create a response and pass instructions parameter
+    instructions = "You are a helpful assistant."
+    response_with_instructions = client.responses.create(
+        model=text_model_id,
+        instructions=instructions,
+        input=messages,
+        stream=False,
+    )
+
+    # Verify we have a valid instructions field
+    assert response_with_instructions.instructions == instructions
+
+    # Finally test instructions parameter with a previous response id
+    instructions2 = "You are a helpful assistant and speak in pirate language."
+    response_with_instructions2 = client.responses.create(
+        model=text_model_id,
+        instructions=instructions2,
+        input=messages,
+        previous_response_id=response_with_instructions.id,
+        stream=False,
+    )
+
+    # Verify instructions from previous response was not carried over to the next response
+    assert response_with_instructions2.instructions == instructions2

tests/integration/common/recordings/models-64a2277c90f0f42576f60c1030e3a020403d34a95f56931b792d5939f4cebc57-abd54ea0.json

@@ -0,0 +1,44 @@
+{
+  "test_id": null,
+  "request": {
+    "method": "POST",
+    "url": "http://0.0.0.0:11434/v1/v1/models",
+    "headers": {},
+    "body": {},
+    "endpoint": "/v1/models",
+    "model": ""
+  },
+  "response": {
+    "body": [
+      {
+        "__type__": "openai.types.model.Model",
+        "__data__": {
+          "id": "llama-guard3:1b",
+          "created": 1753937098,
+          "object": "model",
+          "owned_by": "library"
+        }
+      },
+      {
+        "__type__": "openai.types.model.Model",
+        "__data__": {
+          "id": "all-minilm:l6-v2",
+          "created": 1753936935,
+          "object": "model",
+          "owned_by": "library"
+        }
+      },
+      {
+        "__type__": "openai.types.model.Model",
+        "__data__": {
+          "id": "llama3.2:3b-instruct-fp16",
+          "created": 1753936925,
+          "object": "model",
+          "owned_by": "library"
+        }
+      }
+    ],
+    "is_streaming": false
+  },
+  "id_normalization_mapping": {}
+}

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

@@ -814,6 +814,69 @@ async def test_create_openai_response_with_instructions_and_previous_response(
     assert sent_messages[3].content == "Which is the largest?"
 
 
+async def test_create_openai_response_with_previous_response_instructions(
+    openai_responses_impl, mock_responses_store, mock_inference_api
+):
+    """Test prepending instructions and previous response with instructions."""
+
+    input_item_message = OpenAIResponseMessage(
+        id="123",
+        content="Name some towns in Ireland",
+        role="user",
+    )
+    response_output_message = OpenAIResponseMessage(
+        id="123",
+        content="Galway, Longford, Sligo",
+        status="completed",
+        role="assistant",
+    )
+    response = _OpenAIResponseObjectWithInputAndMessages(
+        created_at=1,
+        id="resp_123",
+        model="fake_model",
+        output=[response_output_message],
+        status="completed",
+        text=OpenAIResponseText(format=OpenAIResponseTextFormat(type="text")),
+        input=[input_item_message],
+        messages=[
+            OpenAIUserMessageParam(content="Name some towns in Ireland"),
+            OpenAIAssistantMessageParam(content="Galway, Longford, Sligo"),
+        ],
+        instructions="You are a helpful assistant.",
+    )
+    mock_responses_store.get_response_object.return_value = response
+
+    model = "meta-llama/Llama-3.1-8B-Instruct"
+    instructions = "You are a geography expert. Provide concise answers."
+
+    mock_inference_api.openai_chat_completion.return_value = fake_stream()
+
+    # Execute
+    await openai_responses_impl.create_openai_response(
+        input="Which is the largest?", model=model, instructions=instructions, previous_response_id="123"
+    )
+
+    # Verify
+    mock_inference_api.openai_chat_completion.assert_called_once()
+    call_args = mock_inference_api.openai_chat_completion.call_args
+    params = call_args.args[0]
+    sent_messages = params.messages
+
+    # Check that instructions were prepended as a system message
+    # and that the previous response instructions were not carried over
+    assert len(sent_messages) == 4, sent_messages
+    assert sent_messages[0].role == "system"
+    assert sent_messages[0].content == instructions
+
+    # Check the rest of the messages were converted correctly
+    assert sent_messages[1].role == "user"
+    assert sent_messages[1].content == "Name some towns in Ireland"
+    assert sent_messages[2].role == "assistant"
+    assert sent_messages[2].content == "Galway, Longford, Sligo"
+    assert sent_messages[3].role == "user"
+    assert sent_messages[3].content == "Which is the largest?"
+
+
 async def test_list_openai_response_input_items_delegation(openai_responses_impl, mock_responses_store):
     """Test that list_openai_response_input_items properly delegates to responses_store with correct parameters."""
     # Setup