mistral[patch]: support model_kwargs (#29838)

- **Description:** Frequency_penalty added as a client parameter
- **Issue:** #29803

---------

Co-authored-by: Chester Curme <chester.curme@gmail.com>

libs/partners/mistralai/langchain_mistralai/chat_models.py

@@ -68,9 +68,10 @@ from langchain_core.output_parsers.openai_tools import (
 from langchain_core.outputs import ChatGeneration, ChatGenerationChunk, ChatResult
 from langchain_core.runnables import Runnable, RunnableMap, RunnablePassthrough
 from langchain_core.tools import BaseTool
-from langchain_core.utils import secret_from_env
+from langchain_core.utils import get_pydantic_field_names, secret_from_env
 from langchain_core.utils.function_calling import convert_to_openai_tool
 from langchain_core.utils.pydantic import is_basemodel_subclass
+from langchain_core.utils.utils import _build_model_kwargs
 from pydantic import (
     BaseModel,
     ConfigDict,
@@ -392,12 +393,22 @@ class ChatMistralAI(BaseChatModel):
     random_seed: Optional[int] = None
     safe_mode: Optional[bool] = None
     streaming: bool = False
+    model_kwargs: Dict[str, Any] = Field(default_factory=dict)
+    """Holds any invocation parameters not explicitly specified."""
 
     model_config = ConfigDict(
         populate_by_name=True,
         arbitrary_types_allowed=True,
     )
 
+    @model_validator(mode="before")
+    @classmethod
+    def build_extra(cls, values: Dict[str, Any]) -> Any:
+        """Build extra kwargs from additional params that were passed in."""
+        all_required_field_names = get_pydantic_field_names(cls)
+        values = _build_model_kwargs(values, all_required_field_names)
+        return values
+
     @property
     def _default_params(self) -> Dict[str, Any]:
         """Get the default parameters for calling the API."""
@@ -408,6 +419,7 @@ class ChatMistralAI(BaseChatModel):
             "top_p": self.top_p,
             "random_seed": self.random_seed,
             "safe_prompt": self.safe_mode,
+            **self.model_kwargs,
         }
         filtered = {k: v for k, v in defaults.items() if v is not None}
         return filtered

libs/partners/mistralai/tests/unit_tests/__snapshots__/test_standard.ambr

@@ -20,6 +20,10 @@
         'type': 'secret',
       }),
       'model': 'mistral-small',
+      'model_kwargs': dict({
+        'stop': list([
+        ]),
+      }),
       'temperature': 0.0,
       'timeout': 60,
       'top_p': 1,

libs/partners/mistralai/tests/unit_tests/test_chat_models.py

@@ -255,3 +255,18 @@ def test_tool_id_conversion() -> None:
     for input_id, expected_output in result_map.items():
         assert _convert_tool_call_id_to_mistral_compatible(input_id) == expected_output
         assert _is_valid_mistral_tool_call_id(expected_output)
+
+
+def test_extra_kwargs() -> None:
+    # Check that foo is saved in extra_kwargs.
+    llm = ChatMistralAI(model="my-model", foo=3, max_tokens=10)  # type: ignore[call-arg]
+    assert llm.max_tokens == 10
+    assert llm.model_kwargs == {"foo": 3}
+
+    # Test that if extra_kwargs are provided, they are added to it.
+    llm = ChatMistralAI(model="my-model", foo=3, model_kwargs={"bar": 2})  # type: ignore[call-arg]
+    assert llm.model_kwargs == {"foo": 3, "bar": 2}
+
+    # Test that if provided twice it errors
+    with pytest.raises(ValueError):
+        ChatMistralAI(model="my-model", foo=3, model_kwargs={"foo": 2})  # type: ignore[call-arg]