phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-10-10 05:24:39 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

test

Browse source 
# What does this PR do?


## Test Plan
This commit is contained in:
Eric Huang 2025-10-08 13:54:19 -07:00
parent 96886afaca
commit 521009048a
207 changed files with 71733 additions and 2042 deletions
Download patch file Download diff file Expand all files Collapse all files

4
llama_stack/providers/utils/inference/litellm_openai_mixin.py
View file

@ -247,6 +247,7 @@ class LiteLLMOpenAIMixin(
guided_choice: list[str] | None = None,
prompt_logprobs: int | None = None,
suffix: str | None = None,
**kwargs: Any,
) -> OpenAICompletion:
model_obj = await self.model_store.get_model(model)
params = await prepare_openai_completion_params(
@ -271,6 +272,7 @@ class LiteLLMOpenAIMixin(
prompt_logprobs=prompt_logprobs,
api_key=self.get_api_key(),
api_base=self.api_base,
**kwargs,
)
return await litellm.atext_completion(**params)
@ -299,6 +301,7 @@ class LiteLLMOpenAIMixin(
top_logprobs: int | None = None,
top_p: float | None = None,
user: str | None = None,
**kwargs: Any,
) -> OpenAIChatCompletion | AsyncIterator[OpenAIChatCompletionChunk]:
# Add usage tracking for streaming when telemetry is active
from llama_stack.providers.utils.telemetry.tracing import get_current_span
@ -335,6 +338,7 @@ class LiteLLMOpenAIMixin(
user=user,
api_key=self.get_api_key(),
api_base=self.api_base,
**kwargs,
)
return await litellm.acompletion(**params)

10
llama_stack/providers/utils/inference/openai_mixin.py
View file

@ -247,6 +247,7 @@ class OpenAIMixin(NeedsRequestProviderData, ABC, BaseModel):
guided_choice: list[str] | None = None,
prompt_logprobs: int | None = None,
suffix: str | None = None,
**kwargs: Any,
) -> OpenAICompletion:
"""
Direct OpenAI completion API call.
@ -261,6 +262,9 @@ class OpenAIMixin(NeedsRequestProviderData, ABC, BaseModel):
if guided_choice:
extra_body["guided_choice"] = guided_choice
# Merge any additional kwargs into extra_body
extra_body.update(kwargs)
# TODO: fix openai_completion to return type compatible with OpenAI's API response
resp = await self.client.completions.create(
**await prepare_openai_completion_params(
@ -313,6 +317,7 @@ class OpenAIMixin(NeedsRequestProviderData, ABC, BaseModel):
top_logprobs: int | None = None,
top_p: float | None = None,
user: str | None = None,
**kwargs: Any,
) -> OpenAIChatCompletion | AsyncIterator[OpenAIChatCompletionChunk]:
"""
Direct OpenAI chat completion API call.
@ -361,7 +366,10 @@ class OpenAIMixin(NeedsRequestProviderData, ABC, BaseModel):
user=user,
)
resp = await self.client.chat.completions.create(**params)
# Pass any additional provider-specific parameters as extra_body
extra_body = kwargs if kwargs else {}
resp = await self.client.chat.completions.create(**params, extra_body=extra_body)
return await self._maybe_overwrite_id(resp, stream) # type: ignore[no-any-return]