Skip to content

Issues: Dao-AILab/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.

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

655 Open 608 Closed
655 Open 608 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

A way to seamlessly replace flash-attention with something to allow for low-data inference on other platforms
#1527 opened Mar 5, 2025 by marinegor
1
Flash Attention 3 compile fails (able to compile 2) : TypeError: _write_ninja_file() got an unexpected keyword argument 'sycl_cflags'
#1524 opened Mar 3, 2025 by FurkanGozukara
13
Which branch is flash attention 3 correct branch?
#1523 opened Mar 3, 2025 by FurkanGozukara
5
How to speed up setup? AMD 9950X 96GB RAM and RTX 5090
#1519 opened Mar 1, 2025 by FurkanGozukara
2
ImportError: /home/data/miniconda3/envs/museTalk/lib/python3.10/site-packages/flash_attn_2_cuda.cpython-310-x86_64-linux-gnu.so: undefined symbol: _ZN3c105ErrorC2ENS_14SourceLocationENSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE
#1518 opened Feb 28, 2025 by gg22mm
1
Inquiry About flash_attn3 b1 usability
#1516 opened Feb 27, 2025 by luyvlei
3
Installing flash-attn & torch with requirements.txt raises build error
#1515 opened Feb 27, 2025 by erenirmak
1
Does it support 2080Ti?
#1514 opened Feb 27, 2025 by Lsnxiaoxiong
3
Does it support RTX 8000?
#1513 opened Feb 27, 2025 by youde2000
1
[QST] Why are Layout<Shape<Int<kNWarps>, _1, _1>> and Tile<Int<16 * kNWarps>, _16, _16>> used in kernel_traits.h for TiledMma?
#1512 opened Feb 27, 2025 by March-H
6
Flash_attn 1.x whl ?
#1508 opened Feb 25, 2025 by darkon12
1
Where is the flash_attn_2_cuda
#1506 opened Feb 24, 2025 by VirgoAsumita
Anyone publishing pre compiled wheels for windows for python 3.10 and 3.11?
#1502 opened Feb 22, 2025 by FurkanGozukara
1
How to adapt code from FA2 to FA3
#1501 opened Feb 21, 2025 by wangchuan
7
No module named 'flash_attn_2_cuda'
#1499 opened Feb 18, 2025 by TheAyes
1
About Triton version of FA
#1496 opened Feb 15, 2025 by YufangMo
2
Missing Requirements
#1494 opened Feb 14, 2025 by domklement
1
Build from source fails with ninja-build installed
#1492 opened Feb 13, 2025 by NorseGaud
Support hdimQK=192, hdimV=128 backward
#1487 opened Feb 10, 2025 by jacob-cursor
1
【RuntimeError: FlashAttention only supports Ampere GPUs or newer. 】on DGX A800 station
#1485 opened Feb 10, 2025 by Alicia320
1
How to Extend FlashAttention to Nearly Infinite HeadDim and Achieve Fully Fused MLA?
#1483 opened Feb 9, 2025 by DefTruth
1 of 3 tasks
2
Can Flash Attention 3 run on A100？
#1481 opened Feb 8, 2025 by wtr0504
10
a segment error while compiling the flash attention v2.7.4 version on the Thor board of the Blackwell architecture
#1479 opened Feb 7, 2025 by wangxianggang1997
FlashAttention-3 is only supported on CUDA 12.3 and above but torch.__version__ = 2.5.0+cu124
#1477 opened Feb 6, 2025 by focusunsink
1
FA v3 UT doesn't run with CUDA 12.6, on H100
#1472 opened Feb 4, 2025 by sunway513
1
ProTip! Adding no:label will show everything without a label.