mirror of
https://github.com/meta-llama/llama-stack.git
synced 2025-07-29 15:23:51 +00:00
undo a typo, add a passthrough distribution
This commit is contained in:
parent
67229f23a4
commit
38fd76f85c
2 changed files with 24 additions and 2 deletions
|
@ -9,7 +9,7 @@ from typing import List, Optional
|
||||||
|
|
||||||
from llama_toolchain.inference.adapters import available_inference_adapters
|
from llama_toolchain.inference.adapters import available_inference_adapters
|
||||||
|
|
||||||
from .datatypes import ApiSurface, Distribution
|
from .datatypes import ApiSurface, Distribution, PassthroughApiAdapter
|
||||||
|
|
||||||
# This is currently duplicated from `requirements.txt` with a few minor changes
|
# This is currently duplicated from `requirements.txt` with a few minor changes
|
||||||
# dev-dependencies like "ufmt" etc. are nuked. A few specialized dependencies
|
# dev-dependencies like "ufmt" etc. are nuked. A few specialized dependencies
|
||||||
|
@ -57,6 +57,28 @@ def available_distributions() -> List[Distribution]:
|
||||||
ApiSurface.inference: inference_adapters_by_id["meta-reference"],
|
ApiSurface.inference: inference_adapters_by_id["meta-reference"],
|
||||||
},
|
},
|
||||||
),
|
),
|
||||||
|
Distribution(
|
||||||
|
name="full-passthrough",
|
||||||
|
description="Point to remote services for all llama stack APIs",
|
||||||
|
additional_pip_packages=[
|
||||||
|
"python-dotenv",
|
||||||
|
"blobfile",
|
||||||
|
"codeshield",
|
||||||
|
"fairscale",
|
||||||
|
"fastapi",
|
||||||
|
"fire",
|
||||||
|
"flake8",
|
||||||
|
"httpx",
|
||||||
|
"huggingface-hub",
|
||||||
|
],
|
||||||
|
adapters={
|
||||||
|
ApiSurface.inference: PassthroughApiAdapter(
|
||||||
|
api_surface=ApiSurface.inference,
|
||||||
|
adapter_id="inference-passthrough",
|
||||||
|
base_url="http://localhost:5001",
|
||||||
|
),
|
||||||
|
},
|
||||||
|
),
|
||||||
Distribution(
|
Distribution(
|
||||||
name="local-ollama",
|
name="local-ollama",
|
||||||
description="Like local-source, but use ollama for running LLM inference",
|
description="Like local-source, but use ollama for running LLM inference",
|
||||||
|
|
|
@ -157,7 +157,7 @@ def create_dynamic_typed_route(func: Any):
|
||||||
)
|
)
|
||||||
|
|
||||||
return StreamingResponse(
|
return StreamingResponse(
|
||||||
sse_generator(func(request2)), media_type="text/event-stream"
|
sse_generator(func(request)), media_type="text/event-stream"
|
||||||
)
|
)
|
||||||
|
|
||||||
else:
|
else:
|
||||||
|
|
Loading…
Add table
Add a link
Reference in a new issue