phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-12-13 04:22:35 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Review, reverse the conditional on api_token_presence

Browse source
This commit is contained in:
Akram Ben Aissi 2025-10-08 20:40:55 +02:00
parent f20eb57bef
commit 24c6b01d16
1 changed files with 2 additions and 6 deletions
Download patch file Download diff file Expand all files Collapse all files

8
llama_stack/providers/remote/inference/vllm/vllm.py
View file

@ -79,18 +79,14 @@ class VLLMInferenceAdapter(OpenAIMixin):
async def check_model_availability(self, model: str) -> bool: async def check_model_availability(self, model: str) -> bool:
""" """
Check model availability only when api_token is configured.
Skip the check when running without authentication. Skip the check when running without authentication.
""" """
if self.config.api_token: if not self.config.api_token:
# If we have a token, perform the normal availability check
try: try:
return model in [m.id async for m in self.client.models.list()] return model in [m.id async for m in self.client.models.list()]
except Exception as e: except Exception as e:
# If listing models fails, log the error but allow the model
log.warning(f"Failed to check model availability: {e}") log.warning(f"Failed to check model availability: {e}")
return True raise ValueError(f"Failed to check model availability: {e}") from e
# Without a token, skip the check to avoid OAuth redirects
return True return True
async def openai_chat_completion( async def openai_chat_completion(