Skip to content

feat[vLLM]: Make VoxtralProcessor compatible with the Transformers modeling backend for vLLM - #47574

Closed
harshaljanjani wants to merge 1 commit into
huggingface:mainfrom
harshaljanjani:feat/voxtral-processor-audio
Closed

feat[vLLM]: Make VoxtralProcessor compatible with the Transformers modeling backend for vLLM#47574
harshaljanjani wants to merge 1 commit into
huggingface:mainfrom
harshaljanjani:feat/voxtral-processor-audio

Conversation

@harshaljanjani

@harshaljanjani harshaljanjani commented Jul 27, 2026

Copy link
Copy Markdown
Contributor

CI

What does this PR do?

→ A blocker to the companion vLLM PR for Voxtral audio generation through the Transformers backend.
VoxtralProcessor.__call__ rejects audio-token text and routes audio only through apply_chat_template. Makes __call__ accept audio like the other audio processors; since MistralCommonBackend does not encode [AUDIO] from text, splices the audio token id into the encoded segments. No modeling changes needed.
Verified no regressions on the Transformers-side in tests/models/voxtral/ (143 non-slow pass; the 5 failing slow integration tests fail identically on main locally, exact-match generation drift, unrelated to this change).
→ I've tried to explain why each non-trivial change was made in a one-line comment given much of this involved unwrapping failure modes layer by layer and uncovering deeper issues in the call chain :)

cc: @eustlb @ebezzam @vasqu

Before (vLLM Voxtral Generation)


After (vLLM Voxtral Generation, After Both PRs)


Code Agent Policy

  • I confirm that this is not a pure code agent PR.

Before submitting

  • This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
  • Did you read the contributor guideline, Pull Request section?
  • Was this discussed/approved via a Github issue or the forum? Please add a link to it if that's the case.
  • Did you make sure to update the documentation with your changes? Here are the documentation guidelines, and here are tips on formatting docstrings.
  • Did you fix any necessary existing tests?

@github-actions

Copy link
Copy Markdown
Contributor

[For maintainers] Suggested jobs to run (before merge)

run-slow: voxtral

@github-actions

Copy link
Copy Markdown
Contributor

CI recap

Dashboard: View test results in Grafana
Latest run: 30254289724
Result: success | Grafana metrics are not available yet.

@hmellor hmellor changed the title feat[vLLM x v5]: Support audio in VoxtralProcessor feat[vLLM]: Make VoxtralProcessor compatible with the Transformers modeling backend for vLLM Jul 27, 2026

@eustlb eustlb left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks a lot for starting this! Nevertheless it's better here to have apply_chat_template rely on __call__ rather than two diff paths (not your at all fault though, was implement like that from the beginning).
Since this is tricky, I am opening a PR to supersede. Will add you as a coauthor there of course! 🤗

@harshaljanjani

Copy link
Copy Markdown
Contributor Author

Thanks for the attribution and makes sense, closing!

@harshaljanjani
harshaljanjani deleted the feat/voxtral-processor-audio branch July 28, 2026 05:11
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants