phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-10-06 20:44:58 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Fixes post-review and split TGI adapter into local and Inference Endpoints ones

Browse source
This commit is contained in:
Celina Hanouti 2024-09-09 17:47:49 +02:00
parent 5ab4fd31f7
commit b96e705680
3 changed files with 98 additions and 17 deletions
Download patch file Download diff file Expand all files Collapse all files

16
llama_toolchain/inference/adapters/tgi/__init__.py
View file

@ -5,14 +5,20 @@
# the root directory of this source tree.
from .config import TGIImplConfig
from .tgi import InferenceEndpointAdapter, LocalTGIAdapter
async def get_adapter_impl(config: TGIImplConfig, _deps):
from .tgi import TGIAdapter
assert isinstance(config, TGIImplConfig), f"Unexpected config type: {type(config)}"
if config.is_local_tgi():
impl = LocalTGIAdapter(config)
elif config.is_inference_endpoint():
impl = InferenceEndpointAdapter(config)
else:
raise ValueError(
"Invalid configuration. Specify either a local URL or Inference Endpoint details."
)
assert isinstance(
config, TGIImplConfig
), f"Unexpected config type: {type(config)}"
impl = TGIAdapter(config)
await impl.initialize()
return impl