feat(langchain,openai): add strict flag to ProviderStrategy structured output (#34149)

libs/langchain_v1/langchain/agents/structured_output.py

@@ -255,21 +255,32 @@ class ProviderStrategy(Generic[SchemaT]):
     def __init__(
         self,
         schema: type[SchemaT],
+        *,
+        strict: bool = False,
     ) -> None:
-        """Initialize ProviderStrategy with schema."""
+        """Initialize ProviderStrategy with schema.
+
+        Args:
+            schema: Schema to enforce via the provider's native structured output.
+            strict: Whether to request strict provider-side schema enforcement.
+        """
         self.schema = schema
-        self.schema_spec = _SchemaSpec(schema)
+        self.schema_spec = _SchemaSpec(schema, strict=strict)
 
     def to_model_kwargs(self) -> dict[str, Any]:
         """Convert to kwargs to bind to a model to force structured output."""
         # OpenAI:
         # - see https://platform.openai.com/docs/guides/structured-outputs
-        response_format = {
+        json_schema: dict[str, Any] = {
+            "name": self.schema_spec.name,
+            "schema": self.schema_spec.json_schema,
+        }
+        if self.schema_spec.strict:
+            json_schema["strict"] = True
+
+        response_format: dict[str, Any] = {
             "type": "json_schema",
-            "json_schema": {
+            "json_schema": json_schema,
-                "name": self.schema_spec.name,
-                "schema": self.schema_spec.json_schema,
-            },
         }
         return {"response_format": response_format}
 

libs/langchain_v1/tests/unit_tests/agents/test_response_format.py

@@ -737,6 +737,18 @@ class TestResponseFormatAsProviderStrategy:
         assert response["structured_response"] == EXPECTED_WEATHER_DICT
         assert len(response["messages"]) == 4
 
+    def test_provider_strategy_strict_flag(self) -> None:
+        """ProviderStrategy should pass through strict flag for provider schemas."""
+        # Default should not set strict
+        strategy_default = ProviderStrategy(WeatherBaseModel)
+        kwargs_default = strategy_default.to_model_kwargs()
+        assert "strict" not in kwargs_default["response_format"]["json_schema"]
+
+        # Explicit strict True should include the flag
+        strategy_strict = ProviderStrategy(WeatherBaseModel, strict=True)
+        kwargs_strict = strategy_strict.to_model_kwargs()
+        assert kwargs_strict["response_format"]["json_schema"]["strict"] is True
+
 
 class TestDynamicModelWithResponseFormat:
     """Test response_format with middleware that modifies the model."""

libs/partners/openai/langchain_openai/chat_models/base.py

@@ -1886,9 +1886,10 @@ class BaseChatOpenAI(BaseChatModel):
             ):
                 # compat with langchain.agents.create_agent response_format, which is
                 # an approximation of OpenAI format
+                strict = response_format["json_schema"].get("strict", None)
                 response_format = cast(dict, response_format["json_schema"]["schema"])
             kwargs["response_format"] = _convert_to_openai_response_format(
-                response_format
+                response_format, strict=strict
             )
         return super().bind(tools=formatted_tools, **kwargs)