Revert "integrations[patch]: remove non-required chat param defaults" (#29048)

Reverts langchain-ai/langchain#26730

discuss best way to release default changes (esp openai temperature)

Author: Erick Friis

libs/partners/anthropic/langchain_anthropic/chat_models.py

@@ -307,7 +307,7 @@ class ChatAnthropic(BaseChatModel):
     Key init args — client params:
         timeout: Optional[float]
             Timeout for requests.
-        max_retries: Optional[int]
+        max_retries: int
             Max number of retries if a request fails.
         api_key: Optional[str]
             Anthropic API key. If not passed in will be read from env var ANTHROPIC_API_KEY.
@@ -558,7 +558,8 @@ class Joke(BaseModel):
     default_request_timeout: Optional[float] = Field(None, alias="timeout")
     """Timeout for requests to Anthropic Completion API."""
 
-    max_retries: Optional[int] = None
+    # sdk default = 2: https://github.com/anthropics/anthropic-sdk-python?tab=readme-ov-file#retries
+    max_retries: int = 2
     """Number of retries allowed for requests sent to the Anthropic Completion API."""
 
     stop_sequences: Optional[List[str]] = Field(None, alias="stop")
@@ -661,10 +662,9 @@ def _client_params(self) -> Dict[str, Any]:
         client_params: Dict[str, Any] = {
             "api_key": self.anthropic_api_key.get_secret_value(),
             "base_url": self.anthropic_api_url,
+            "max_retries": self.max_retries,
             "default_headers": (self.default_headers or None),
         }
-        if self.max_retries is not None:
-            client_params["max_retries"] = self.max_retries
         # value <= 0 indicates the param should be ignored. None is a meaningful value
         # for Anthropic client and treated differently than not specifying the param at
         # all.

libs/partners/fireworks/langchain_fireworks/chat_models.py

@@ -316,7 +316,7 @@ def is_lc_serializable(cls) -> bool:
         default="accounts/fireworks/models/mixtral-8x7b-instruct", alias="model"
     )
     """Model name to use."""
-    temperature: Optional[float] = None
+    temperature: float = 0.0
     """What sampling temperature to use."""
     stop: Optional[Union[str, List[str]]] = Field(default=None, alias="stop_sequences")
     """Default stop sequences."""

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

@@ -22,7 +22,6 @@
       'request_timeout': 60.0,
       'stop': list([
       ]),
-      'temperature': 0.0,
     }),
     'lc': 1,
     'name': 'ChatFireworks',

libs/partners/groq/langchain_groq/chat_models.py

@@ -119,7 +119,7 @@ class ChatGroq(BaseChatModel):
     Key init args — client params:
         timeout: Union[float, Tuple[float, float], Any, None]
             Timeout for requests.
-        max_retries: Optional[int]
+        max_retries: int
             Max number of retries.
         api_key: Optional[str]
             Groq API key. If not passed in will be read from env var GROQ_API_KEY.
@@ -303,7 +303,7 @@ class Joke(BaseModel):
     async_client: Any = Field(default=None, exclude=True)  #: :meta private:
     model_name: str = Field(default="mixtral-8x7b-32768", alias="model")
     """Model name to use."""
-    temperature: Optional[float] = None
+    temperature: float = 0.7
     """What sampling temperature to use."""
     stop: Optional[Union[List[str], str]] = Field(default=None, alias="stop_sequences")
     """Default stop sequences."""
@@ -327,11 +327,11 @@ class Joke(BaseModel):
     )
     """Timeout for requests to Groq completion API. Can be float, httpx.Timeout or
         None."""
-    max_retries: Optional[int] = None
+    max_retries: int = 2
     """Maximum number of retries to make when generating."""
     streaming: bool = False
     """Whether to stream the results or not."""
-    n: Optional[int] = None
+    n: int = 1
     """Number of chat completions to generate for each prompt."""
     max_tokens: Optional[int] = None
     """Maximum number of tokens to generate."""
@@ -379,11 +379,10 @@ def build_extra(cls, values: Dict[str, Any]) -> Any:
     @model_validator(mode="after")
     def validate_environment(self) -> Self:
         """Validate that api key and python package exists in environment."""
-        if self.n is not None and self.n < 1:
+        if self.n < 1:
             raise ValueError("n must be at least 1.")
-        elif self.n is not None and self.n > 1 and self.streaming:
+        if self.n > 1 and self.streaming:
             raise ValueError("n must be 1 when streaming.")
-
         if self.temperature == 0:
             self.temperature = 1e-8
 
@@ -393,11 +392,10 @@ def validate_environment(self) -> Self:
             ),
             "base_url": self.groq_api_base,
             "timeout": self.request_timeout,
+            "max_retries": self.max_retries,
             "default_headers": self.default_headers,
             "default_query": self.default_query,
         }
-        if self.max_retries is not None:
-            client_params["max_retries"] = self.max_retries
 
         try:
             import groq

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

@@ -17,6 +17,7 @@
       'max_retries': 2,
       'max_tokens': 100,
       'model_name': 'mixtral-8x7b-32768',
+      'n': 1,
       'request_timeout': 60.0,
       'stop': list([
       ]),

libs/partners/mistralai/langchain_mistralai/chat_models.py

@@ -95,11 +95,8 @@ def _create_retry_decorator(
     """Returns a tenacity retry decorator, preconfigured to handle exceptions"""
 
     errors = [httpx.RequestError, httpx.StreamError]
-    kwargs: dict = dict(
-        error_types=errors, max_retries=llm.max_retries, run_manager=run_manager
-    )
     return create_base_retry_decorator(
-        **{k: v for k, v in kwargs.items() if v is not None}
+        error_types=errors, max_retries=llm.max_retries, run_manager=run_manager
     )
 
 
@@ -383,13 +380,13 @@ class ChatMistralAI(BaseChatModel):
         default_factory=secret_from_env("MISTRAL_API_KEY", default=None),
     )
     endpoint: Optional[str] = Field(default=None, alias="base_url")
-    max_retries: Optional[int] = None
-    timeout: Optional[int] = None
-    max_concurrent_requests: Optional[int] = None
+    max_retries: int = 5
+    timeout: int = 120
+    max_concurrent_requests: int = 64
     model: str = Field(default="mistral-small", alias="model_name")
-    temperature: Optional[float] = None
+    temperature: float = 0.7
     max_tokens: Optional[int] = None
-    top_p: Optional[float] = None
+    top_p: float = 1
     """Decode using nucleus sampling: consider the smallest set of tokens whose
        probability sum is at least top_p. Must be in the closed interval [0.0, 1.0]."""
     random_seed: Optional[int] = None

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

@@ -9,6 +9,7 @@
     ]),
     'kwargs': dict({
       'endpoint': 'boo',
+      'max_concurrent_requests': 64,
       'max_retries': 2,
       'max_tokens': 100,
       'mistral_api_key': dict({
@@ -21,6 +22,7 @@
       'model': 'mistral-small',
       'temperature': 0.0,
       'timeout': 60,
+      'top_p': 1,
     }),
     'lc': 1,
     'name': 'ChatMistralAI',

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

@@ -79,7 +79,7 @@ class AzureChatOpenAI(BaseChatOpenAI):
             https://learn.microsoft.com/en-us/azure/ai-services/openai/reference#rest-api-versioning
         timeout: Union[float, Tuple[float, float], Any, None]
             Timeout for requests.
-        max_retries: Optional[int]
+        max_retries: int
             Max number of retries.
         organization: Optional[str]
             OpenAI organization ID. If not passed in will be read from env
@@ -586,9 +586,9 @@ def is_lc_serializable(cls) -> bool:
     @model_validator(mode="after")
     def validate_environment(self) -> Self:
         """Validate that api key and python package exists in environment."""
-        if self.n is not None and self.n < 1:
+        if self.n < 1:
             raise ValueError("n must be at least 1.")
-        elif self.n is not None and self.n > 1 and self.streaming:
+        if self.n > 1 and self.streaming:
             raise ValueError("n must be 1 when streaming.")
 
         if self.disabled_params is None:
@@ -641,11 +641,10 @@ def validate_environment(self) -> Self:
             "organization": self.openai_organization,
             "base_url": self.openai_api_base,
             "timeout": self.request_timeout,
+            "max_retries": self.max_retries,
             "default_headers": self.default_headers,
             "default_query": self.default_query,
         }
-        if self.max_retries is not None:
-            client_params["max_retries"] = self.max_retries
         if not self.client:
             sync_specific = {"http_client": self.http_client}
             self.root_client = openai.AzureOpenAI(**client_params, **sync_specific)  # type: ignore[arg-type]

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

@@ -414,7 +414,7 @@ class BaseChatOpenAI(BaseChatModel):
     root_async_client: Any = Field(default=None, exclude=True)  #: :meta private:
     model_name: str = Field(default="gpt-3.5-turbo", alias="model")
     """Model name to use."""
-    temperature: Optional[float] = None
+    temperature: float = 0.7
     """What sampling temperature to use."""
     model_kwargs: Dict[str, Any] = Field(default_factory=dict)
     """Holds any model parameters valid for `create` call not explicitly specified."""
@@ -435,7 +435,7 @@ class BaseChatOpenAI(BaseChatModel):
     )
     """Timeout for requests to OpenAI completion API. Can be float, httpx.Timeout or 
         None."""
-    max_retries: Optional[int] = None
+    max_retries: int = 2
     """Maximum number of retries to make when generating."""
     presence_penalty: Optional[float] = None
     """Penalizes repeated tokens."""
@@ -453,7 +453,7 @@ class BaseChatOpenAI(BaseChatModel):
     """Modify the likelihood of specified tokens appearing in the completion."""
     streaming: bool = False
     """Whether to stream the results or not."""
-    n: Optional[int] = None
+    n: int = 1
     """Number of chat completions to generate for each prompt."""
     top_p: Optional[float] = None
     """Total probability mass of tokens to consider at each step."""
@@ -537,9 +537,9 @@ def validate_temperature(cls, values: Dict[str, Any]) -> Any:
     @model_validator(mode="after")
     def validate_environment(self) -> Self:
         """Validate that api key and python package exists in environment."""
-        if self.n is not None and self.n < 1:
+        if self.n < 1:
             raise ValueError("n must be at least 1.")
-        elif self.n is not None and self.n > 1 and self.streaming:
+        if self.n > 1 and self.streaming:
             raise ValueError("n must be 1 when streaming.")
 
         # Check OPENAI_ORGANIZATION for backwards compatibility.
@@ -556,12 +556,10 @@ def validate_environment(self) -> Self:
             "organization": self.openai_organization,
             "base_url": self.openai_api_base,
             "timeout": self.request_timeout,
+            "max_retries": self.max_retries,
             "default_headers": self.default_headers,
             "default_query": self.default_query,
         }
-        if self.max_retries is not None:
-            client_params["max_retries"] = self.max_retries
-
         if self.openai_proxy and (self.http_client or self.http_async_client):
             openai_proxy = self.openai_proxy
             http_client = self.http_client
@@ -616,14 +614,14 @@ def _default_params(self) -> Dict[str, Any]:
             "stop": self.stop or None,  # also exclude empty list for this
             "max_tokens": self.max_tokens,
             "extra_body": self.extra_body,
-            "n": self.n,
-            "temperature": self.temperature,
             "reasoning_effort": self.reasoning_effort,
         }
 
         params = {
             "model": self.model_name,
             "stream": self.streaming,
+            "n": self.n,
+            "temperature": self.temperature,
             **{k: v for k, v in exclude_if_none.items() if v is not None},
             **self.model_kwargs,
         }
@@ -1575,7 +1573,7 @@ class ChatOpenAI(BaseChatOpenAI):  # type: ignore[override]
 
         timeout: Union[float, Tuple[float, float], Any, None]
             Timeout for requests.
-        max_retries: Optional[int]
+        max_retries: int
             Max number of retries.
         api_key: Optional[str]
             OpenAI API key. If not passed in will be read from env var OPENAI_API_KEY.

libs/partners/openai/tests/unit_tests/chat_models/__snapshots__/test_azure_standard.ambr

@@ -15,6 +15,7 @@
       }),
       'max_retries': 2,
       'max_tokens': 100,
+      'n': 1,
       'openai_api_key': dict({
         'id': list([
           'AZURE_OPENAI_API_KEY',