chore(litellm): bump dev image to v1.88.0-rc.1 (fixes chatgpt Responses output bug)#460
Merged
Merged
Conversation
…es output bug) Codex calls (openai-codex/gpt-5.4*) were silently falling through to the OpenRouter Nemotron fallback because LiteLLM's chatgpt/ provider threw `'ResponsesAPIResponse' object has no attribute 'output'` on every real ChatGPT response. Agents posted (via Nemotron) so the failure was invisible — but the dev tier was never running on Codex output despite paying for two ChatGPT accounts. Verified live 2026-05-31: - v1.83.7-stable (prior): fails with the ChatgptException above. - v1.86.2 (latest stable that drops -stable suffix): SAME bug — fix not in v1.86.x. - v1.88.0-rc.1: openai-codex/gpt-5.4-mini returns real Codex output (model field "openai-codex/gpt-5.4-mini", not Nemotron). Pinning to the RC because v1.87.0 stable isn't published on ghcr yet (only RCs available). Revisit when v1.87.0 or v1.88.0 stable lands. No prod values change — values.yaml still on v1.82.3-stable; only dev gets the RC. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Contributor
Author
|
Squash-merged to main per feedback-pr-merge-pattern. |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
Bumps LiteLLM dev image from
v1.83.7-stabletov1.88.0-rc.1to fix the upstream'ResponsesAPIResponse' object has no attribute 'output'bug in LiteLLM'schatgpt/provider.Why
Codex calls (
openai-codex/gpt-5.4*) silently fell through to the OpenRouter Nemotron fallback because LiteLLM threw aChatgptExceptionon every real ChatGPT response. Agents kept posting (via Nemotron) so the failure was invisible — but the dev tier was never actually running on Codex output despite paying for two ChatGPT accounts.Verified live 2026-05-31 via direct
kubectl set imagetest:model: openai-codex/gpt-5.4-miniTrade-off
v1.88.0-rc.1is a release candidate. v1.87.0 stable isn't published on ghcr.io yet (only RCs available). Pin will be revisited when a non-RC tag with the fix lands.Test plan
openai-codex/gpt-5.4-minireturns real Codex output (not fallback).kubectl set image— verified stable for 5+ min before this PR was written.helm upgradekeeps the image pinned atv1.88.0-rc.1instead of reverting tov1.83.7-stable.🤖 Generated with Claude Code