Downloads · 30 days
0
flash-attention-windows-wheel
lldacing/flash-attention-windows-wheel
flash-attention-windows-wheel is a machine learning model from lldacing. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as bsd-3-clause.
Access
Public
Updated May 31, 2025
Repo size
4.6 GB
Likes
333
Public
Repo makeup
Click a slice to open those files.
.whl4.2 GB · 100%
At a glance
- License
- bsd-3-clause
- Access
- Public
- Created
- Nov 15, 2024
- Updated
- May 31, 2025
- SHA
- b894b89d
Model card
From the Hugging Face model README
Windows wheels of flash-attention
Build cuda wheel steps
- First clone code
git clone https://github.com/Dao-AILab/flash-attention
cd flash-attention
- Switch tag branch, such as
v2.7.0.post2(you can get latest tag bygit describe --tagsor list all available tags bygit tag -l)
git checkout -b v2.7.0.post2 v2.7.0.post2
-
Download WindowsWhlBuilder_cuda.bat into
flash-attention -
To build with MSVC, please open the "Native Tools Command Prompt for Visual Studio". The exact name may depend on your version of Windows, Visual Studio, and cpu architecture (in my case it was "x64 Native Tools Command Prompt for VS 2022".)
My Visual Studio Installer version
-
Switch python env and make sure the corresponding torch cuda version is installed
-
Start task
# Build with 1 parallel workers (I used 8 workers on i9-14900KF-3.20GHz-RAM64G, which took about 30 minutes.)
# If you want reset workers, you should edit `WindowsWhlBuilder_cuda.bat` and modify `set MAX_JOBS=1`.(I tried to modify it by parameters, but failed)
WindowsWhlBuilder_cuda.bat
# Build with sm80 and sm120
WindowsWhlBuilder_cuda.bat CUDA_ARCH="80;120"
# Enable cxx11abi
WindowsWhlBuilder_cuda.bat CUDA_ARCH="80;120" FORCE_CXX11_ABI=TRUE
- Wheel file will be placed in the
distdirectory
- License
- bsd-3-clause
- Created
- Nov 15, 2024
- Updated
- May 31, 2025