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

Adapter -> Provider

Browse source
This commit is contained in:
Ashwin Bharambe 2024-08-05 13:26:29 -07:00
parent db3e6dda07
commit 65a9e40174
15 changed files with 119 additions and 110 deletions
Download patch file Download diff file Expand all files Collapse all files

6
llama_toolchain/inference/inference.py
View file

@ -11,7 +11,7 @@ from typing import AsyncIterator, Dict, Union
from llama_models.llama3_1.api.datatypes import StopReason
from llama_models.sku_list import resolve_model
from llama_toolchain.distribution.datatypes import Adapter, Api
from llama_toolchain.distribution.datatypes import Api, ProviderSpec
from .api.config import MetaReferenceImplConfig
from .api.datatypes import (
@ -29,7 +29,9 @@ from .api.endpoints import (
from .model_parallel import LlamaModelParallelGenerator
async def get_adapter_impl(config: MetaReferenceImplConfig, _deps: Dict[Api, Adapter]):
async def get_provider_impl(
config: MetaReferenceImplConfig, _deps: Dict[Api, ProviderSpec]
):
assert isinstance(
config, MetaReferenceImplConfig
), f"Unexpected config type: {type(config)}"

2
llama_toolchain/inference/ollama.py
View file

@ -37,7 +37,7 @@ from .api.endpoints import (
)
def get_adapter_impl(config: OllamaImplConfig) -> Inference:
def get_provider_impl(config: OllamaImplConfig) -> Inference:
assert isinstance(
config, OllamaImplConfig
), f"Unexpected config type: {type(config)}"

12
llama_toolchain/inference/adapters.py → llama_toolchain/inference/providers.py
View file

@ -6,14 +6,14 @@
from typing import List
from llama_toolchain.distribution.datatypes import Adapter, Api, SourceAdapter
from llama_toolchain.distribution.datatypes import Api, InlineProviderSpec, ProviderSpec
def available_inference_adapters() -> List[Adapter]:
def available_inference_providers() -> List[ProviderSpec]:
return [
SourceAdapter(
InlineProviderSpec(
api=Api.inference,
adapter_id="meta-reference",
provider_id="meta-reference",
pip_packages=[
"torch",
"zmq",
@ -21,9 +21,9 @@ def available_inference_adapters() -> List[Adapter]:
module="llama_toolchain.inference.inference",
config_class="llama_toolchain.inference.inference.MetaReferenceImplConfig",
),
SourceAdapter(
InlineProviderSpec(
api=Api.inference,
adapter_id="meta-ollama",
provider_id="meta-ollama",
pip_packages=[
"ollama",
],