Skip to content

select float32 for krea2 denoise on mps - #9579

Open
grunblatt-git wants to merge 1 commit into
invoke-ai:mainfrom
grunblatt-git:krea2-select-float32-for-mps
Open

select float32 for krea2 denoise on mps#9579
grunblatt-git wants to merge 1 commit into
invoke-ai:mainfrom
grunblatt-git:krea2-select-float32-for-mps

Conversation

@grunblatt-git

Copy link
Copy Markdown

Summary

This is a Performance Improvement for Krea2 on MPS devices where float32 outperforms bfloat16.

For other devices the previous behavior remains unchanged if precision: auto is set in the config. Otherwise the custom precision setting is honored for krea2 inference.

Related Issues / Discussions

Closes #9575

Checklist

  • The PR has a short but descriptive title, suitable for a changelog
  • Tests added / updated (if applicable)
  • ❗Changes to a redux slice have a corresponding migration
  • Documentation added / updated (if applicable)
  • Updated What's New copy (if doing a release after this PR)

@github-actions github-actions Bot added python PRs that change python files backend PRs that change backend files labels Sep 8, 2026
@lstein lstein added the 7.0.0 label Sep 11, 2026
@lstein lstein moved this to 7.0 Theme: Tabbed Layout UI in Invoke - Community Roadmap Sep 11, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

7.0.0 backend PRs that change backend files python PRs that change python files

Projects

Status: 7.0 Theme: Tabbed Layout UI

Development

Successfully merging this pull request may close these issues.

[bug]: Krea2 Inference is slow on MPS (due to hardcoded bfloat16 dtype)

3 participants