phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-10-12 13:57:57 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

featu: support passing "extra body" throught to providers

Browse source 
# What does this PR do?
Allows passing through extra_body parameters to inference providers.


closes #2720

## Test Plan
CI and added new test
This commit is contained in:
Eric Huang 2025-10-10 16:00:30 -07:00
parent 80d58ab519
commit c4dbaa9d4c
41 changed files with 3145 additions and 200 deletions
Download patch file Download diff file Expand all files Collapse all files

8
llama_stack/providers/remote/inference/passthrough/passthrough.py
View file

@ -13,9 +13,9 @@ from llama_stack.apis.inference import (
Inference,
OpenAIChatCompletion,
OpenAIChatCompletionChunk,
OpenAIChatCompletionRequest,
OpenAIChatCompletionRequestWithExtraBody,
OpenAICompletion,
OpenAICompletionRequest,
OpenAICompletionRequestWithExtraBody,
OpenAIEmbeddingsResponse,
)
from llama_stack.apis.models import Model
@ -79,7 +79,7 @@ class PassthroughInferenceAdapter(Inference):
async def openai_completion(
self,
params: OpenAICompletionRequest,
params: OpenAICompletionRequestWithExtraBody,
) -> OpenAICompletion:
client = self._get_client()
model_obj = await self.model_store.get_model(params.model)
@ -93,7 +93,7 @@ class PassthroughInferenceAdapter(Inference):
async def openai_chat_completion(
self,
params: OpenAIChatCompletionRequest,
params: OpenAIChatCompletionRequestWithExtraBody,
) -> OpenAIChatCompletion | AsyncIterator[OpenAIChatCompletionChunk]:
client = self._get_client()
model_obj = await self.model_store.get_model(params.model)