Skip to content
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

[Feature Request] Support AMX BF16 #19937

Open
jchia opened this issue Mar 15, 2024 · 0 comments
Open

[Feature Request] Support AMX BF16 #19937

jchia opened this issue Mar 15, 2024 · 0 comments
Assignees
@yufenglee
Labels
feature request request for unsupported feature or enhancement quantization issues related to quantization

Comments

@jchia
Copy link

jchia commented Mar 15, 2024 •
edited
Loading

Describe the feature request

#14042 Adds AMX support for int8 GEMM. It would be good to also have support for BF16.

If there is AMX support for BF16 GEMM, should the BF16 AMX be automatically used when the inputs are FP16 or FP32 or does the user need to explicitly Cast?

Describe scenario use case

Higher-performance for BF16 GEMM on CPUs with AMX BF16.
@jchia jchia added the feature request request for unsupported feature or enhancement label Mar 15, 2024
@snnn snnn added the quantization issues related to quantization label Mar 19, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@yufenglee yufenglee

Labels
feature request request for unsupported feature or enhancement quantization issues related to quantization
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@jchia @snnn @yufenglee