Adding run_id to ModelRequest and ModelResponse

docs/agents.md

@@ -320,7 +320,8 @@ async def main():
                         content='What is the capital of France?',
                         timestamp=datetime.datetime(...),
                     )
-                ]
+                ],
+                run_id='...',
             )
         ),
         CallToolsNode(
@@ -329,6 +330,7 @@ async def main():
                 usage=RequestUsage(input_tokens=56, output_tokens=7),
                 model_name='gpt-5',
                 timestamp=datetime.datetime(...),
+                run_id='...',
             )
         ),
         End(data=FinalResult(output='The capital of France is Paris.')),
@@ -382,7 +384,8 @@ async def main():
                             content='What is the capital of France?',
                             timestamp=datetime.datetime(...),
                         )
-                    ]
+                    ],
+                    run_id='...',
                 )
             ),
             CallToolsNode(
@@ -391,6 +394,7 @@ async def main():
                     usage=RequestUsage(input_tokens=56, output_tokens=7),
                     model_name='gpt-5',
                     timestamp=datetime.datetime(...),
+                    run_id='...',
                 )
             ),
             End(data=FinalResult(output='The capital of France is Paris.')),
@@ -1044,7 +1048,8 @@ with capture_run_messages() as messages:  # (2)!
                         content='Please get me the volume of a box with size 6.',
                         timestamp=datetime.datetime(...),
                     )
-                ]
+                ],
+                run_id='...',
             ),
             ModelResponse(
                 parts=[
@@ -1057,6 +1062,7 @@ with capture_run_messages() as messages:  # (2)!
                 usage=RequestUsage(input_tokens=62, output_tokens=4),
                 model_name='gpt-5',
                 timestamp=datetime.datetime(...),
+                run_id='...',
             ),
             ModelRequest(
                 parts=[
@@ -1066,7 +1072,8 @@ with capture_run_messages() as messages:  # (2)!
                         tool_call_id='pyd_ai_tool_call_id',
                         timestamp=datetime.datetime(...),
                     )
-                ]
+                ],
+                run_id='...',
             ),
             ModelResponse(
                 parts=[
@@ -1079,6 +1086,7 @@ with capture_run_messages() as messages:  # (2)!
                 usage=RequestUsage(input_tokens=72, output_tokens=8),
                 model_name='gpt-5',
                 timestamp=datetime.datetime(...),
+                run_id='...',
             ),
         ]
         """

docs/api/models/function.md

@@ -29,7 +29,8 @@ async def model_function(
                     content='Testing my agent...',
                     timestamp=datetime.datetime(...),
                 )
-            ]
+            ],
+            run_id='...',
         )
     ]
     """

docs/deferred-tools.md

@@ -106,7 +106,8 @@ print(result.all_messages())
                 content='Delete `__init__.py`, write `Hello, world!` to `README.md`, and clear `.env`',
                 timestamp=datetime.datetime(...),
             )
-        ]
+        ],
+        run_id='...',
     ),
     ModelResponse(
         parts=[
@@ -129,6 +130,7 @@ print(result.all_messages())
         usage=RequestUsage(input_tokens=63, output_tokens=21),
         model_name='gpt-5',
         timestamp=datetime.datetime(...),
+        run_id='...',
     ),
     ModelRequest(
         parts=[
@@ -138,7 +140,8 @@ print(result.all_messages())
                 tool_call_id='update_file_readme',
                 timestamp=datetime.datetime(...),
             )
-        ]
+        ],
+        run_id='...',
     ),
     ModelRequest(
         parts=[
@@ -154,7 +157,8 @@ print(result.all_messages())
                 tool_call_id='delete_file',
                 timestamp=datetime.datetime(...),
             ),
-        ]
+        ],
+        run_id='...',
     ),
     ModelResponse(
         parts=[
@@ -165,6 +169,7 @@ print(result.all_messages())
         usage=RequestUsage(input_tokens=79, output_tokens=39),
         model_name='gpt-5',
         timestamp=datetime.datetime(...),
+        run_id='...',
     ),
 ]
 """
@@ -275,7 +280,8 @@ async def main():
                     content='Calculate the answer to the ultimate question of life, the universe, and everything',
                     timestamp=datetime.datetime(...),
                 )
-            ]
+            ],
+            run_id='...',
         ),
         ModelResponse(
             parts=[
@@ -290,6 +296,7 @@ async def main():
             usage=RequestUsage(input_tokens=63, output_tokens=13),
             model_name='gpt-5',
             timestamp=datetime.datetime(...),
+            run_id='...',
         ),
         ModelRequest(
             parts=[
@@ -299,7 +306,8 @@ async def main():
                     tool_call_id='pyd_ai_tool_call_id',
                     timestamp=datetime.datetime(...),
                 )
-            ]
+            ],
+            run_id='...',
         ),
         ModelResponse(
             parts=[
@@ -310,6 +318,7 @@ async def main():
             usage=RequestUsage(input_tokens=64, output_tokens=28),
             model_name='gpt-5',
             timestamp=datetime.datetime(...),
+            run_id='...',
         ),
     ]
     """

docs/durable_execution/temporal.md

@@ -172,7 +172,7 @@ As workflows and activities run in separate processes, any values passed between
 
 To account for these limitations, tool functions and the [event stream handler](#streaming) running inside activities receive a limited version of the agent's [`RunContext`][pydantic_ai.tools.RunContext], and it's your responsibility to make sure that the [dependencies](../dependencies.md) object provided to [`TemporalAgent.run()`][pydantic_ai.durable_exec.temporal.TemporalAgent.run] can be serialized using Pydantic.
 
-Specifically, only the `deps`, `retries`, `tool_call_id`, `tool_name`, `tool_call_approved`, `retry`, `max_retries`, `run_step` and `partial_output` fields are available by default, and trying to access `model`, `usage`, `prompt`, `messages`, or `tracer` will raise an error.
+Specifically, only the `deps`, `run_id`, `retries`, `tool_call_id`, `tool_name`, `tool_call_approved`, `retry`, `max_retries`, `run_step` and `partial_output` fields are available by default, and trying to access `model`, `usage`, `prompt`, `messages`, or `tracer` will raise an error.
 If you need one or more of these attributes to be available inside activities, you can create a [`TemporalRunContext`][pydantic_ai.durable_exec.temporal.TemporalRunContext] subclass with custom `serialize_run_context` and `deserialize_run_context` class methods and pass it to [`TemporalAgent`][pydantic_ai.durable_exec.temporal.TemporalAgent] as `run_context_type`.
 
 ### Streaming

docs/message-history.md

@@ -50,7 +50,8 @@ print(result.all_messages())
                 content='Tell me a joke.',
                 timestamp=datetime.datetime(...),
             ),
-        ]
+        ],
+        run_id='...',
     ),
     ModelResponse(
         parts=[
@@ -61,6 +62,7 @@ print(result.all_messages())
         usage=RequestUsage(input_tokens=60, output_tokens=12),
         model_name='gpt-5',
         timestamp=datetime.datetime(...),
+        run_id='...',
     ),
 ]
 """
@@ -92,7 +94,8 @@ async def main():
                         content='Tell me a joke.',
                         timestamp=datetime.datetime(...),
                     ),
-                ]
+                ],
+                run_id='...',
             )
         ]
         """
@@ -118,7 +121,8 @@ async def main():
                         content='Tell me a joke.',
                         timestamp=datetime.datetime(...),
                     ),
-                ]
+                ],
+                run_id='...',
             ),
             ModelResponse(
                 parts=[
@@ -129,6 +133,7 @@ async def main():
                 usage=RequestUsage(input_tokens=50, output_tokens=12),
                 model_name='gpt-5',
                 timestamp=datetime.datetime(...),
+                run_id='...',
             ),
         ]
         """
@@ -172,7 +177,8 @@ print(result2.all_messages())
                 content='Tell me a joke.',
                 timestamp=datetime.datetime(...),
             ),
-        ]
+        ],
+        run_id='...',
     ),
     ModelResponse(
         parts=[
@@ -183,14 +189,16 @@ print(result2.all_messages())
         usage=RequestUsage(input_tokens=60, output_tokens=12),
         model_name='gpt-5',
         timestamp=datetime.datetime(...),
+        run_id='...',
     ),
     ModelRequest(
         parts=[
             UserPromptPart(
                 content='Explain?',
                 timestamp=datetime.datetime(...),
             )
-        ]
+        ],
+        run_id='...',
     ),
     ModelResponse(
         parts=[
@@ -201,6 +209,7 @@ print(result2.all_messages())
         usage=RequestUsage(input_tokens=61, output_tokens=26),
         model_name='gpt-5',
         timestamp=datetime.datetime(...),
+        run_id='...',
     ),
 ]
 """
@@ -293,7 +302,8 @@ print(result2.all_messages())
                 content='Tell me a joke.',
                 timestamp=datetime.datetime(...),
             ),
-        ]
+        ],
+        run_id='...',
     ),
     ModelResponse(
         parts=[
@@ -304,14 +314,16 @@ print(result2.all_messages())
         usage=RequestUsage(input_tokens=60, output_tokens=12),
         model_name='gpt-5',
         timestamp=datetime.datetime(...),
+        run_id='...',
     ),
     ModelRequest(
         parts=[
             UserPromptPart(
                 content='Explain?',
                 timestamp=datetime.datetime(...),
             )
-        ]
+        ],
+        run_id='...',
     ),
     ModelResponse(
         parts=[
@@ -322,6 +334,7 @@ print(result2.all_messages())
         usage=RequestUsage(input_tokens=61, output_tokens=26),
         model_name='gemini-2.5-pro',
         timestamp=datetime.datetime(...),
+        run_id='...',
     ),
 ]
 """

docs/testing.md

@@ -127,7 +127,8 @@ async def test_forecast():
                     content='What will the weather be like in London on 2024-11-28?',
                     timestamp=IsNow(tz=timezone.utc),  # (7)!
                 ),
-            ]
+            ],
+            run_id=IsStr(),
         ),
         ModelResponse(
             parts=[
@@ -146,6 +147,7 @@ async def test_forecast():
             ),
             model_name='test',
             timestamp=IsNow(tz=timezone.utc),
+            run_id=IsStr(),
         ),
         ModelRequest(
             parts=[
@@ -156,6 +158,7 @@ async def test_forecast():
                     timestamp=IsNow(tz=timezone.utc),
                 ),
             ],
+            run_id=IsStr(),
         ),
         ModelResponse(
             parts=[
@@ -169,6 +172,7 @@ async def test_forecast():
             ),
             model_name='test',
             timestamp=IsNow(tz=timezone.utc),
+            run_id=IsStr(),
         ),
     ]
 ```

docs/tools.md

@@ -87,7 +87,8 @@ print(dice_result.all_messages())
                 content='My guess is 4',
                 timestamp=datetime.datetime(...),
             ),
-        ]
+        ],
+        run_id='...',
     ),
     ModelResponse(
         parts=[
@@ -98,6 +99,7 @@ print(dice_result.all_messages())
         usage=RequestUsage(input_tokens=90, output_tokens=2),
         model_name='gemini-2.5-flash',
         timestamp=datetime.datetime(...),
+        run_id='...',
     ),
     ModelRequest(
         parts=[
@@ -107,7 +109,8 @@ print(dice_result.all_messages())
                 tool_call_id='pyd_ai_tool_call_id',
                 timestamp=datetime.datetime(...),
             )
-        ]
+        ],
+        run_id='...',
     ),
     ModelResponse(
         parts=[
@@ -118,6 +121,7 @@ print(dice_result.all_messages())
         usage=RequestUsage(input_tokens=91, output_tokens=4),
         model_name='gemini-2.5-flash',
         timestamp=datetime.datetime(...),
+        run_id='...',
     ),
     ModelRequest(
         parts=[
@@ -127,7 +131,8 @@ print(dice_result.all_messages())
                 tool_call_id='pyd_ai_tool_call_id',
                 timestamp=datetime.datetime(...),
             )
-        ]
+        ],
+        run_id='...',
     ),
     ModelResponse(
         parts=[
@@ -138,6 +143,7 @@ print(dice_result.all_messages())
         usage=RequestUsage(input_tokens=92, output_tokens=12),
         model_name='gemini-2.5-flash',
         timestamp=datetime.datetime(...),
+        run_id='...',
     ),
 ]
 """