phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-06-27 18:50:41 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Add provider deprecation support; change directory structure (#397)

Browse source 
* Add provider deprecation support; change directory structure

* fix a couple dangling imports

* move the meta_reference safety dir also
This commit is contained in:
Ashwin Bharambe 2024-11-07 13:04:53 -08:00 • committed by GitHub
parent 36e2538eb0
commit 694c142b89
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
58 changed files with 61 additions and 120 deletions
Download patch file Download diff file Expand all files Collapse all files

26
llama_stack/providers/registry/inference.py
View file

@ -27,8 +27,8 @@ def available_providers() -> List[ProviderSpec]:
api=Api.inference,
provider_type="meta-reference",
pip_packages=META_REFERENCE_DEPS,
module="llama_stack.providers.inline.meta_reference.inference",
config_class="llama_stack.providers.inline.meta_reference.inference.MetaReferenceInferenceConfig",
module="llama_stack.providers.inline.inference.meta_reference",
config_class="llama_stack.providers.inline.inference.meta_reference.MetaReferenceInferenceConfig",
),
InlineProviderSpec(
api=Api.inference,
@ -40,8 +40,17 @@ def available_providers() -> List[ProviderSpec]:
"torchao==0.5.0",
]
),
module="llama_stack.providers.inline.meta_reference.inference",
config_class="llama_stack.providers.inline.meta_reference.inference.MetaReferenceQuantizedInferenceConfig",
module="llama_stack.providers.inline.inference.meta_reference",
config_class="llama_stack.providers.inline.inference.meta_reference.MetaReferenceQuantizedInferenceConfig",
),
InlineProviderSpec(
api=Api.inference,
provider_type="vllm",
pip_packages=[
"vllm",
],
module="llama_stack.providers.inline.inference.vllm",
config_class="llama_stack.providers.inline.inference.vllm.VLLMConfig",
),
remote_provider_spec(
api=Api.inference,
@ -140,13 +149,4 @@ def available_providers() -> List[ProviderSpec]:
config_class="llama_stack.providers.remote.inference.databricks.DatabricksImplConfig",
),
),
InlineProviderSpec(
api=Api.inference,
provider_type="vllm",
pip_packages=[
"vllm",
],
module="llama_stack.providers.inline.vllm",
config_class="llama_stack.providers.inline.vllm.VLLMConfig",
),
]