Skip to content

Fix zero-copy numpy arrays dtype mismatch - #4731

Open
michalk8 wants to merge 1 commit into
ml-explore:mainfrom
michalk8:feature/fix-zero-copy-dtype
Open

michalk8 wants to merge 1 commit into
ml-explore:mainfrom
michalk8:feature/fix-zero-copy-dtype

Conversation

@michalk8

Copy link
Copy Markdown
Collaborator
  • ☑️ I understand it is strictly prohibited to use AI to write PR description
  • AI usage disclosure: Tests

Zero-copy from numpy array was only looking at the width of the array's dtype and the MLX dtype, leading to incorrect interpretation of the data (e.g. int32 could be read as fp32); this PR fixes it.

@michalk8
michalk8 requested a review from zcbenz October 20, 2026 07:52

This branch has not been deployed

No deployments
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.

1 participant