phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-07-27 22:48:51 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

feat(starter)!: simplify starter distro; litellm model registry changes (#2916)

Browse source
This commit is contained in:
Ashwin Bharambe 2025-07-25 15:02:04 -07:00 • committed by GitHub
parent 3344d8a9e5
commit 9583f468f8
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
64 changed files with 2027 additions and 4092 deletions
Download patch file Download diff file Expand all files Collapse all files

13
llama_stack/providers/remote/inference/vllm/vllm.py
View file

@ -299,7 +299,10 @@ class VLLMInferenceAdapter(Inference, ModelsProtocolPrivate):
self.client = None
async def initialize(self) -> None:
pass
if not self.config.url:
raise ValueError(
"You must provide a URL in run.yaml (or via the VLLM_URL environment variable) to use vLLM."
)
async def should_refresh_models(self) -> bool:
return self.config.refresh_models
@ -337,9 +340,6 @@ class VLLMInferenceAdapter(Inference, ModelsProtocolPrivate):
HealthResponse: A dictionary containing the health status.
"""
try:
if not self.config.url:
return HealthResponse(status=HealthStatus.ERROR, message="vLLM URL is not set")
client = self._create_client() if self.client is None else self.client
_ = [m async for m in client.models.list()] # Ensure the client is initialized
return HealthResponse(status=HealthStatus.OK)
@ -355,11 +355,6 @@ class VLLMInferenceAdapter(Inference, ModelsProtocolPrivate):
if self.client is not None:
return
if not self.config.url:
raise ValueError(
"You must provide a vLLM URL in the run.yaml file (or set the VLLM_URL environment variable)"
)
log.info(f"Initializing vLLM client with base_url={self.config.url}")
self.client = self._create_client()