phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-10-07 20:50:52 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

instantiate inference models

Browse source
This commit is contained in:
Xi Yan 2024-09-19 21:33:35 -07:00
parent d2ec822b12
commit 7071c46422
6 changed files with 40 additions and 20 deletions
Download patch file Download diff file Expand all files Collapse all files

6
llama_stack/distribution/server/server.py
View file

@ -321,12 +321,10 @@ async def resolve_impls(
inner_specs=inner_specs,
)
for k, v in specs.items():
cprint(k, "blue")
cprint(v, "blue")
sorted_specs = topological_sort(specs.values())
cprint(f"sorted_specs={sorted_specs}", "red")
impls = {}
for spec in sorted_specs:
api = spec.api

3
llama_stack/distribution/utils/dynamic.py
View file

@ -53,7 +53,8 @@ async def instantiate_provider(
args = [inner_impls, deps]
elif isinstance(provider_config, str) and provider_config == "models-router":
config = None
args = [[], deps]
assert len(deps) == 1 and Api.models in deps
args = [deps[Api.models]]
else:
raise ValueError(f"provider_config {provider_config} is not valid")
else: