Added request_id to the model response. (#1547)

peterHoburg · web-flow · commit b7137cb455ee · 2025-05-19T13:38:56.000-06:00
diff --git a/.gitignore b/.gitignore
@@ -17,3 +17,5 @@ examples/pydantic_ai_examples/.chat_app_messages.sqlite
 /docs-site/.wrangler/
 /CLAUDE.md
 node_modules/
+**.idea/
+.coverage*
diff --git a/docs/agents.md b/docs/agents.md
@@ -155,6 +155,7 @@ async def main():
                 model_name='gpt-4o',
                 timestamp=datetime.datetime(...),
                 kind='response',
+                vendor_id=None,
             )
         ),
         End(data=FinalResult(output='Paris', tool_name=None, tool_call_id=None)),
@@ -226,6 +227,7 @@ async def main():
                     model_name='gpt-4o',
                     timestamp=datetime.datetime(...),
                     kind='response',
+                    vendor_id=None,
                 )
             ),
             End(data=FinalResult(output='Paris', tool_name=None, tool_call_id=None)),
@@ -829,6 +831,7 @@ with capture_run_messages() as messages:  # (2)!
                 model_name='gpt-4o',
                 timestamp=datetime.datetime(...),
                 kind='response',
+                vendor_id=None,
             ),
             ModelRequest(
                 parts=[
@@ -862,6 +865,7 @@ with capture_run_messages() as messages:  # (2)!
                 model_name='gpt-4o',
                 timestamp=datetime.datetime(...),
                 kind='response',
+                vendor_id=None,
             ),
         ]
         """
diff --git a/docs/direct.md b/docs/direct.md
@@ -95,6 +95,7 @@ async def main():
         model_name='gpt-4.1-nano',
         timestamp=datetime.datetime(...),
         kind='response',
+        vendor_id=None,
     )
     """
 ```
diff --git a/docs/message-history.md b/docs/message-history.md
@@ -74,6 +74,7 @@ print(result.all_messages())
         model_name='gpt-4o',
         timestamp=datetime.datetime(...),
         kind='response',
+        vendor_id=None,
     ),
 ]
 """
@@ -159,6 +160,7 @@ async def main():
                 model_name='gpt-4o',
                 timestamp=datetime.datetime(...),
                 kind='response',
+                vendor_id=None,
             ),
         ]
         """
@@ -225,6 +227,7 @@ print(result2.all_messages())
         model_name='gpt-4o',
         timestamp=datetime.datetime(...),
         kind='response',
+        vendor_id=None,
     ),
     ModelRequest(
         parts=[
@@ -254,6 +257,7 @@ print(result2.all_messages())
         model_name='gpt-4o',
         timestamp=datetime.datetime(...),
         kind='response',
+        vendor_id=None,
     ),
 ]
 """
@@ -367,6 +371,7 @@ print(result2.all_messages())
         model_name='gpt-4o',
         timestamp=datetime.datetime(...),
         kind='response',
+        vendor_id=None,
     ),
     ModelRequest(
         parts=[
@@ -396,6 +401,7 @@ print(result2.all_messages())
         model_name='gemini-1.5-pro',
         timestamp=datetime.datetime(...),
         kind='response',
+        vendor_id=None,
     ),
 ]
 """
diff --git a/docs/models/index.md b/docs/models/index.md
@@ -105,6 +105,7 @@ print(response.all_messages())
         model_name='claude-3-5-sonnet-latest',
         timestamp=datetime.datetime(...),
         kind='response',
+        vendor_id=None,
     ),
 ]
 """
diff --git a/docs/tools.md b/docs/tools.md
@@ -106,6 +106,7 @@ print(dice_result.all_messages())
         model_name='gemini-1.5-flash',
         timestamp=datetime.datetime(...),
         kind='response',
+        vendor_id=None,
     ),
     ModelRequest(
         parts=[
@@ -139,6 +140,7 @@ print(dice_result.all_messages())
         model_name='gemini-1.5-flash',
         timestamp=datetime.datetime(...),
         kind='response',
+        vendor_id=None,
     ),
     ModelRequest(
         parts=[
@@ -170,6 +172,7 @@ print(dice_result.all_messages())
         model_name='gemini-1.5-flash',
         timestamp=datetime.datetime(...),
         kind='response',
+        vendor_id=None,
     ),
 ]
 """
diff --git a/pydantic_ai_slim/pydantic_ai/agent.py b/pydantic_ai_slim/pydantic_ai/agent.py
@@ -585,6 +585,7 @@ async def main():
                         model_name='gpt-4o',
                         timestamp=datetime.datetime(...),
                         kind='response',
+                        vendor_id=None,
                     )
                 ),
                 End(data=FinalResult(output='Paris', tool_name=None, tool_call_id=None)),
@@ -1854,6 +1855,7 @@ async def main():
                     model_name='gpt-4o',
                     timestamp=datetime.datetime(...),
                     kind='response',
+                    vendor_id=None,
                 )
             ),
             End(data=FinalResult(output='Paris', tool_name=None, tool_call_id=None)),
@@ -1999,6 +2001,7 @@ async def main():
                             model_name='gpt-4o',
                             timestamp=datetime.datetime(...),
                             kind='response',
+                            vendor_id=None,
                         )
                     ),
                     End(data=FinalResult(output='Paris', tool_name=None, tool_call_id=None)),
diff --git a/pydantic_ai_slim/pydantic_ai/direct.py b/pydantic_ai_slim/pydantic_ai/direct.py
@@ -52,6 +52,7 @@ async def main():
             model_name='claude-3-5-haiku-latest',
             timestamp=datetime.datetime(...),
             kind='response',
+            vendor_id=None,
         )
         '''
     ```
@@ -108,6 +109,7 @@ def model_request_sync(
         model_name='claude-3-5-haiku-latest',
         timestamp=datetime.datetime(...),
         kind='response',
+        vendor_id=None,
     )
     '''
     ```
diff --git a/pydantic_ai_slim/pydantic_ai/messages.py b/pydantic_ai_slim/pydantic_ai/messages.py
@@ -569,6 +569,9 @@ class ModelResponse:
     For OpenAI models, this may include 'logprobs', 'finish_reason', etc.
     """
 
+    vendor_id: str | None = None
+    """Vendor ID as specified by the model provider. This can be used to track the specific request to the model."""
+
     def otel_events(self) -> list[Event]:
         """Return OpenTelemetry events for the response."""
         result: list[Event] = []
diff --git a/pydantic_ai_slim/pydantic_ai/models/anthropic.py b/pydantic_ai_slim/pydantic_ai/models/anthropic.py
@@ -262,7 +262,7 @@ def _process_response(self, response: AnthropicMessage) -> ModelResponse:
                     )
                 )
 
-        return ModelResponse(items, usage=_map_usage(response), model_name=response.model)
+        return ModelResponse(items, usage=_map_usage(response), model_name=response.model, vendor_id=response.id)
 
     async def _process_streamed_response(self, response: AsyncStream[RawMessageStreamEvent]) -> StreamedResponse:
         peekable_response = _utils.PeekableAsyncStream(response)
diff --git a/pydantic_ai_slim/pydantic_ai/models/bedrock.py b/pydantic_ai_slim/pydantic_ai/models/bedrock.py
@@ -271,7 +271,8 @@ async def _process_response(self, response: ConverseResponseTypeDef) -> ModelRes
             response_tokens=response['usage']['outputTokens'],
             total_tokens=response['usage']['totalTokens'],
         )
-        return ModelResponse(items, usage=u, model_name=self.model_name)
+        vendor_id = response.get('ResponseMetadata', {}).get('RequestId', None)
+        return ModelResponse(items, usage=u, model_name=self.model_name, vendor_id=vendor_id)
 
     @overload
     async def _messages_create(
diff --git a/pydantic_ai_slim/pydantic_ai/models/groq.py b/pydantic_ai_slim/pydantic_ai/models/groq.py
@@ -239,7 +239,9 @@ def _process_response(self, response: chat.ChatCompletion) -> ModelResponse:
         if choice.message.tool_calls is not None:
             for c in choice.message.tool_calls:
                 items.append(ToolCallPart(tool_name=c.function.name, args=c.function.arguments, tool_call_id=c.id))
-        return ModelResponse(items, usage=_map_usage(response), model_name=response.model, timestamp=timestamp)
+        return ModelResponse(
+            items, usage=_map_usage(response), model_name=response.model, timestamp=timestamp, vendor_id=response.id
+        )
 
     async def _process_streamed_response(self, response: AsyncStream[chat.ChatCompletionChunk]) -> GroqStreamedResponse:
         """Process a streamed response, and prepare a streaming response to return."""
diff --git a/pydantic_ai_slim/pydantic_ai/models/mistral.py b/pydantic_ai_slim/pydantic_ai/models/mistral.py
@@ -325,7 +325,9 @@ def _process_response(self, response: MistralChatCompletionResponse) -> ModelRes
                 tool = self._map_mistral_to_pydantic_tool_call(tool_call=tool_call)
                 parts.append(tool)
 
-        return ModelResponse(parts, usage=_map_usage(response), model_name=response.model, timestamp=timestamp)
+        return ModelResponse(
+            parts, usage=_map_usage(response), model_name=response.model, timestamp=timestamp, vendor_id=response.id
+        )
 
     async def _process_streamed_response(
         self,
diff --git a/pydantic_ai_slim/pydantic_ai/models/openai.py b/pydantic_ai_slim/pydantic_ai/models/openai.py
@@ -339,6 +339,7 @@ def _process_response(self, response: chat.ChatCompletion) -> ModelResponse:
             model_name=response.model,
             timestamp=timestamp,
             vendor_details=vendor_details,
+            vendor_id=response.id,
         )
 
     async def _process_streamed_response(self, response: AsyncStream[ChatCompletionChunk]) -> OpenAIStreamedResponse:
diff --git a/tests/models/test_anthropic.py b/tests/models/test_anthropic.py
@@ -184,6 +184,7 @@ async def test_sync_request_text_response(allow_model_requests: None):
                 ),
                 model_name='claude-3-5-haiku-123',
                 timestamp=IsNow(tz=timezone.utc),
+                vendor_id='123',
             ),
             ModelRequest(parts=[UserPromptPart(content='hello', timestamp=IsNow(tz=timezone.utc))]),
             ModelResponse(
@@ -197,6 +198,7 @@ async def test_sync_request_text_response(allow_model_requests: None):
                 ),
                 model_name='claude-3-5-haiku-123',
                 timestamp=IsNow(tz=timezone.utc),
+                vendor_id='123',
             ),
         ]
     )
@@ -287,6 +289,7 @@ async def test_request_structured_response(allow_model_requests: None):
                 ),
                 model_name='claude-3-5-haiku-123',
                 timestamp=IsNow(tz=timezone.utc),
+                vendor_id='123',
             ),
             ModelRequest(
                 parts=[
@@ -356,6 +359,7 @@ async def get_location(loc_name: str) -> str:
                 ),
                 model_name='claude-3-5-haiku-123',
                 timestamp=IsNow(tz=timezone.utc),
+                vendor_id='123',
             ),
             ModelRequest(
                 parts=[
@@ -384,6 +388,7 @@ async def get_location(loc_name: str) -> str:
                 ),
                 model_name='claude-3-5-haiku-123',
                 timestamp=IsNow(tz=timezone.utc),
+                vendor_id='123',
             ),
             ModelRequest(
                 parts=[
@@ -406,6 +411,7 @@ async def get_location(loc_name: str) -> str:
                 ),
                 model_name='claude-3-5-haiku-123',
                 timestamp=IsNow(tz=timezone.utc),
+                vendor_id='123',
             ),
         ]
     )
@@ -754,6 +760,7 @@ async def get_image() -> BinaryContent:
                 ),
                 model_name='claude-3-5-sonnet-20241022',
                 timestamp=IsDatetime(),
+                vendor_id='msg_01BPu4UTHXhqtR1TvsRhBLYY',
             ),
             ModelRequest(
                 parts=[
@@ -792,6 +799,7 @@ async def get_image() -> BinaryContent:
                 ),
                 model_name='claude-3-5-sonnet-20241022',
                 timestamp=IsDatetime(),
+                vendor_id='msg_01Ua6uyZUF15YV3G1PusaqSq',
             ),
         ]
     )
@@ -924,6 +932,7 @@ def simple_instructions():
                 ),
                 model_name='claude-3-opus-20240229',
                 timestamp=IsDatetime(),
+                vendor_id='msg_01U58nruzfn9BrXrrF2hhb4m',
             ),
         ]
     )
diff --git a/tests/models/test_groq.py b/tests/models/test_groq.py
@@ -147,13 +147,15 @@ async def test_request_simple_success(allow_model_requests: None):
                 usage=Usage(requests=1),
                 model_name='llama-3.3-70b-versatile-123',
                 timestamp=datetime(2024, 1, 1, 0, 0, tzinfo=timezone.utc),
+                vendor_id='123',
             ),
             ModelRequest(parts=[UserPromptPart(content='hello', timestamp=IsNow(tz=timezone.utc))]),
             ModelResponse(
                 parts=[TextPart(content='world')],
                 usage=Usage(requests=1),
                 model_name='llama-3.3-70b-versatile-123',
                 timestamp=datetime(2024, 1, 1, 0, 0, tzinfo=timezone.utc),
+                vendor_id='123',
             ),
         ]
     )
@@ -206,6 +208,7 @@ async def test_request_structured_response(allow_model_requests: None):
                 usage=Usage(requests=1),
                 model_name='llama-3.3-70b-versatile-123',
                 timestamp=datetime(2024, 1, 1, tzinfo=timezone.utc),
+                vendor_id='123',
             ),
             ModelRequest(
                 parts=[
@@ -293,6 +296,7 @@ async def get_location(loc_name: str) -> str:
                 usage=Usage(requests=1, request_tokens=2, response_tokens=1, total_tokens=3),
                 model_name='llama-3.3-70b-versatile-123',
                 timestamp=datetime(2024, 1, 1, 0, 0, tzinfo=timezone.utc),
+                vendor_id='123',
             ),
             ModelRequest(
                 parts=[
@@ -315,6 +319,7 @@ async def get_location(loc_name: str) -> str:
                 usage=Usage(requests=1, request_tokens=3, response_tokens=2, total_tokens=6),
                 model_name='llama-3.3-70b-versatile-123',
                 timestamp=datetime(2024, 1, 1, 0, 0, tzinfo=timezone.utc),
+                vendor_id='123',
             ),
             ModelRequest(
                 parts=[
@@ -331,6 +336,7 @@ async def get_location(loc_name: str) -> str:
                 usage=Usage(requests=1),
                 model_name='llama-3.3-70b-versatile-123',
                 timestamp=datetime(2024, 1, 1, 0, 0, tzinfo=timezone.utc),
+                vendor_id='123',
             ),
         ]
     )
@@ -567,6 +573,7 @@ async def get_image() -> BinaryContent:
                 usage=Usage(requests=1, request_tokens=192, response_tokens=8, total_tokens=200),
                 model_name='meta-llama/llama-4-scout-17b-16e-instruct',
                 timestamp=IsDatetime(),
+                vendor_id='chatcmpl-3c327c89-e9f5-4aac-a5d5-190e6f6f25c9',
             ),
             ModelRequest(
                 parts=[
@@ -590,6 +597,7 @@ async def get_image() -> BinaryContent:
                 usage=Usage(requests=1, request_tokens=2552, response_tokens=11, total_tokens=2563),
                 model_name='meta-llama/llama-4-scout-17b-16e-instruct',
                 timestamp=IsDatetime(),
+                vendor_id='chatcmpl-82dfad42-6a28-4089-82c3-c8633f626c0d',
             ),
         ]
     )
@@ -669,6 +677,7 @@ async def test_groq_model_instructions(allow_model_requests: None, groq_api_key:
                 usage=Usage(requests=1, request_tokens=48, response_tokens=8, total_tokens=56),
                 model_name='llama-3.3-70b-versatile',
                 timestamp=IsDatetime(),
+                vendor_id='chatcmpl-7586b6a9-fb4b-4ec7-86a0-59f0a77844cf',
             ),
         ]
     )
diff --git a/tests/models/test_mistral.py b/tests/models/test_mistral.py
diff --git a/tests/models/test_openai.py b/tests/models/test_openai.py
diff --git a/tests/test_agent.py b/tests/test_agent.py
diff --git a/tests/test_mcp.py b/tests/test_mcp.py

Original file line number	Diff line number	Diff line change
`@@ -95,6 +95,7 @@ async def main():`
`95`	`95`	`model_name='gpt-4.1-nano',`
`96`	`96`	`timestamp=datetime.datetime(...),`
`97`	`97`	`kind='response',`
	`98`	`+ vendor_id=None,`
`98`	`99`	`)`
`99`	`100`	`"""`
`100`	`101`	```
Original file line number	Diff line number	Diff line change
`@@ -105,6 +105,7 @@ print(response.all_messages())`
`105`	`105`	`model_name='claude-3-5-sonnet-latest',`
`106`	`106`	`timestamp=datetime.datetime(...),`
`107`	`107`	`kind='response',`
	`108`	`+ vendor_id=None,`
`108`	`109`	`),`
`109`	`110`	`]`
`110`	`111`	`"""`
Original file line number	Diff line number	Diff line change
`@@ -262,7 +262,7 @@ def _process_response(self, response: AnthropicMessage) -> ModelResponse:`
`262`	`262`	`)`
`263`	`263`	`)`
`264`	`264`
`265`		`- return ModelResponse(items, usage=_map_usage(response), model_name=response.model)`
	`265`	`+ return ModelResponse(items, usage=_map_usage(response), model_name=response.model, vendor_id=response.id)`
`266`	`266`
`267`	`267`	`async def _process_streamed_response(self, response: AsyncStream[RawMessageStreamEvent]) -> StreamedResponse:`
`268`	`268`	`peekable_response = _utils.PeekableAsyncStream(response)`