phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-12-14 22:52:39 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

feat: Add allow_listing_models

Browse source 
• Add allow_listing_models configuration flag to VLLM provider to control model listing behavior
• Implement allow_listing_models() method across all providers with default implementations in base classes
• Prevent HTTP requests to /v1/models endpoint when allow_listing_models=false for improved security and performance
• Fix unit tests to include allow_listing_models method in test classes and mock objects
This commit is contained in:
Akram Ben Aissi 2025-10-04 00:17:53 +02:00
parent 188a56af5c
commit e9214f9004
15 changed files with 143 additions and 25 deletions
Download patch file Download diff file Expand all files Collapse all files

6
llama_stack/core/routing_tables/models.py
View file

@ -43,6 +43,12 @@ class ModelsRoutingTable(CommonRoutingTableImpl, Models):
await self.update_registered_models(provider_id, models)
async def list_models(self) -> ListModelsResponse:
# Check if providers allow listing models before returning models
for provider_id, provider in self.impls_by_provider_id.items():
allow_listing_models = await provider.allow_listing_models()
logger.debug(f"Provider {provider_id}: allow_listing_models={allow_listing_models}")
if not allow_listing_models:
logger.debug(f"Provider {provider_id} has allow_listing_models disabled")
return ListModelsResponse(data=await self.get_all_with_type("model"))
async def openai_list_models(self) -> OpenAIListModelsResponse: