pytorch / pytorch/executorch

Fix CPU performance on Whisper (index_put)

Open
#17,420 0 comments 0 reactions 1 assignee View on GitHub

@GregoryComer is already working on this.

Since Feb 12, 2026.

Dominant language
Python
Stars
5k
Forks
1.2k
Avg merge
2d 10h
Merged PRs (30d)
581

Description

Whisper performance on CPU (through Optimum) is very slow - less than 1 token/s in decode for medium and larger. This is because it's using functional index_put, which is very slow. We should be using the KV cache update logic.

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.