Skip to content

Fallback to FP16 from BF16 when MPS is available (for Apple Silicon) #3

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Open
wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Fallback to FP16 from BF16 when MPS is available (for Apple Silicon) #3

wants to merge 1 commit into from

Conversation

cchance27
Copy link

  • Added checks for using mps device instead of just cuda.
  • On MPS force models to fp16 instead of AMP Autocast as it's currently not supported on Apple Silicon.

It works on my Macbook M3, using ~19ish gigs similar to BF16, and works well if a bit slow.

Before merging please confirm i didn't break anything on cuda/cpu, as i don't have a pc to test on for that.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
1 participant
@cchance27