[Frontend][OpenAI] Support for returning max_model_len on /v1/models response (#4643)

vllm/entrypoints/openai/protocol.py

@@ -82,6 +82,7 @@ class ModelCard(OpenAIBaseModel):
     owned_by: str = "vllm"
     root: Optional[str] = None
     parent: Optional[str] = None
+    max_model_len: Optional[int] = None
     permission: List[ModelPermission] = Field(default_factory=list)
 
 

vllm/entrypoints/openai/serving_engine.py

@@ -62,6 +62,7 @@ class OpenAIServing:
         """Show available models. Right now we only have one model."""
         model_cards = [
             ModelCard(id=served_model_name,
+                      max_model_len=self.max_model_len,
                       root=self.served_model_names[0],
                       permission=[ModelPermission()])
             for served_model_name in self.served_model_names