phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-07-29 07:14:20 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Improve TGI adapter initialization condition

Browse source
This commit is contained in:
Celina Hanouti 2024-09-10 18:22:09 +02:00
parent 2ac8e7b901
commit 0964b0a74a
2 changed files with 2 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

2
llama_toolchain/inference/adapters/tgi/__init__.py
View file

@ -11,7 +11,7 @@ from .tgi import InferenceEndpointAdapter, TGIAdapter
async def get_adapter_impl(config: TGIImplConfig, _deps):
assert isinstance(config, TGIImplConfig), f"Unexpected config type: {type(config)}"
if config.is_local_tgi():
if config.url is not None:
impl = TGIAdapter(config)
elif config.is_inference_endpoint():
impl = InferenceEndpointAdapter(config)

4
llama_toolchain/inference/adapters/tgi/config.py
View file

@ -7,6 +7,7 @@
from typing import Optional
from huggingface_hub import HfApi
from llama_models.schema_utils import json_schema_type
from pydantic import BaseModel, Field
@ -31,6 +32,3 @@ class TGIImplConfig(BaseModel):
def get_namespace(self) -> str:
return HfApi().whoami()["name"]
def is_local_tgi(self) -> bool:
return self.url is not None and self.url.startswith("http://localhost")