set default max_tokens to 1024

langchain-ai · May 8, 2024 · a791549 · a791549
1 parent 014e87b
commit a791549
Showing 1 changed file with 1 addition and 1 deletion.
diff --git a/libs/ai-endpoints/langchain_nvidia_ai_endpoints/chat_models.py b/libs/ai-endpoints/langchain_nvidia_ai_endpoints/chat_models.py
@@ -134,7 +134,7 @@ class ChatNVIDIA(nvidia_ai_endpoints._NVIDIAClient, BaseChatModel):
     infer_endpoint: str = Field("{base_url}/chat/completions")
     model: str = Field(_default_model, description="Name of the model to invoke")
     temperature: Optional[float] = Field(description="Sampling temperature in [0, 1]")
-    max_tokens: Optional[int] = Field(description="Maximum # of tokens to generate")
+    max_tokens: Optional[int] = Field(1024, description="Maximum # of tokens to generate")
     top_p: Optional[float] = Field(description="Top-p for distribution sampling")
     seed: Optional[int] = Field(description="The seed for deterministic results")
     bad: Optional[Sequence[str]] = Field(description="Bad words to avoid (cased)")