forked from Dao-AILab/flash-attention
-
Notifications
You must be signed in to change notification settings - Fork 50
Issues: ROCm/flash-attention
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.
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
Author
Label
Projects
Milestones
Assignee
Sort
Issues list
[Issue]: Test failing with ROCm 6.3.1 on MI250X
Under Investigation
#120
opened Jan 29, 2025 by
al-rigazzi
Issue: GFX1151 fails to build. Undeclared identifier 'CK_TILE_BUFFER_RESOURCE_3RD_DWORD'
Under Investigation
#111
opened Dec 17, 2024 by
kiram9
[Issue]: is scaled_dot_product_attention part of flash attention?
#79
opened Sep 2, 2024 by
unclemusclez
[Feature]: Flash Attention 3 Support for MI300X GPUs
Feature Request
Under Investigation
#71
opened Aug 1, 2024 by
codinggosu
[Feature]: Support for newer flash-attention versions (e.g. ≥2.1.0)
Feature Request
Under Investigation
#53
opened May 22, 2024 by
JiahuaZhao
[Issue]: RuntimeError: Expected dout_seq_stride == out_seq_stride to be true, but got false.
#41
opened Feb 3, 2024 by
donglixp
[Issue]: Expected dout_seq_stride == out_seq_stride to be true, but got false
#40
opened Jan 30, 2024 by
ehartford
Support for other modules (rotary, xentropy, layer_norm)
function
#34
opened Jan 15, 2024 by
bbartoldson
replace kernel implementation using CK tile-programming performant kernels
#33
opened Jan 10, 2024 by
carlushuang
4 tasks
ProTip!
Add no:assignee to see everything that’s not assigned.