phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-10-04 04:04:14 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Bring agentic system api to toolchain

Browse source 
Add adapter dependencies and resolve adapters using a topological sort
This commit is contained in:
Ashwin Bharambe 2024-08-04 10:53:38 -07:00
parent b0e5340645
commit be19b22391
31 changed files with 2740 additions and 25 deletions
Download patch file Download diff file Expand all files Collapse all files

4
llama_toolchain/inference/client.py
View file

@ -23,6 +23,10 @@ from .api import (
from .event_logger import EventLogger
async def get_client_impl(base_url: str):
return InferenceClient(base_url)
class InferenceClient(Inference):
def __init__(self, base_url: str):
print(f"Initializing client for {base_url}")

8
llama_toolchain/inference/inference.py
View file

@ -6,11 +6,13 @@
import asyncio
from typing import AsyncIterator, Union
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, ApiSurface
from .api.config import MetaReferenceImplConfig
from .api.datatypes import (
ChatCompletionResponseEvent,
@ -27,7 +29,9 @@ from .api.endpoints import (
from .model_parallel import LlamaModelParallelGenerator
async def get_adapter_impl(config: MetaReferenceImplConfig):
async def get_adapter_impl(
config: MetaReferenceImplConfig, _deps: Dict[ApiSurface, Adapter]
):
assert isinstance(
config, MetaReferenceImplConfig
), f"Unexpected config type: {type(config)}"