Add LiteLLM provider into responses_api_models#990
Open
imxj wants to merge 2 commits intoNVIDIA-NeMo:mainfrom
Open
Add LiteLLM provider into responses_api_models#990imxj wants to merge 2 commits intoNVIDIA-NeMo:mainfrom
imxj wants to merge 2 commits intoNVIDIA-NeMo:mainfrom
Conversation
imxj
force-pushed
the
feat/support_litellm_responses
branch
2 times, most recently
from
805aaef to
f0558e5
Compare
imxj
changed the title
LiteLLM proxies may return chat.completion format or hybrid response objects when called via /v1/responses. This normalizes those responses to the expected Responses API shape so downstream NeMoGymResponse validation succeeds. Also fixes reasoning.effort="none" validation error for native Responses API responses. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> Signed-off-by: Jin Xu <jinx@nvidia.com>
Revert openai_model/app.py to its clean state and move the LiteLLM response normalization logic into a new responses_api_models/litellm_model that extends SimpleModelServer. This keeps proxy-specific workarounds (reasoning.effort="none" fix, chat.completion->response normalization) isolated from the native OpenAI model server. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> Signed-off-by: Jin Xu <jinx@nvidia.com>
imxj
force-pushed
the
feat/support_litellm_responses
branch
from
105e7c6 to
307b580
Compare
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
responses_api_models/litellm_modelserver for LiteLLM proxy endpointsLiteLLMModelServerextendsopenai_model'sSimpleModelServer, overriding onlyresponses()to normalize LiteLLM proxy quirkschat.completionhybrid responses to standard Responses API format (LiteLLM may downgrade/v1/responsesinternally)reasoning.effort="none"(string) tonullPydantic validation error for native Responses API responsesopenai_modelremains clean -- no proxy-specific workaroundsTest