flash-attention pre-build wheels
This repository provides wheels for the pre-built flash-attention.
Since building flash-attention takes a very long time and is resource-intensive,
I also build and provide combinations of CUDA and PyTorch that are not officially distributed.
The building Github Actions Workflow can be found here.
The built packages are available on the release page.
This repository uses a self-hosted runner and AWS CodeBuild for building the wheels. If you find this project helpful, please consider sponsoring to help maintain the infrastructure!

Table of Contents
Install
- Select the versions for Python, CUDA, PyTorch, and flash_attn.
-
Find the corresponding version of a wheel from the below Package section and releases
-
Direct Install or Download and Local Install
Self build
If you cannot find the version you are looking for, you can fork this repository and create a wheel on GitHub Actions.
- Fork this repository
- Edit workflow file
.github/workflows/build.yml to set the version you want to build.
- Add tag
v*.*.* to trigger the build workflow.
Please note that depending on the combination of versions, it may not be possible to build.
Self-Hosted Runner Build
In some version combinations, you cannot build wheels on GitHub-hosted runners due to job time limitations.
To build the wheels for these versions, you can use self-hosted runners.
Edit env file to set the environment variables.
Edit compose.yml file if you use repository folked from this repository.
Then, build and run the docker container.
Packages
Linux x86_64
Flash-Attention 2.8.3
Packages for Flash-Attention 2.8.3
Flash-Attention 2.8.2
Packages for Flash-Attention 2.8.2
Flash-Attention 2.8.1
Packages for Flash-Attention 2.8.1
Flash-Attention 2.8.0
Packages for Flash-Attention 2.8.0
Flash-Attention 2.7.4
Packages for Flash-Attention 2.7.4
Flash-Attention 2.7.4.post1
Packages for Flash-Attention 2.7.4.post1
| Python |
PyTorch |
CUDA |
package |
| 3.12 |
2.7.0 |
12.6.3 |
Release1 |
| 3.12 |
2.7.0 |
12.4.1 |
Release1 |
| 3.12 |
2.7.0 |
11.8.0 |
Release1 |
| 3.12 |
2.6.0 |
12.6.3 |
Release1, Release2, Release3 |
| 3.12 |
2.6.0 |
12.4.1 |
Release1, Release2, Release3 |
| 3.12 |
2.6.0 |
11.8.0 |
Release1 |
| 3.12 |
2.5.1 |
12.6.3 |
Release1, Release2, Release3 |
| 3.12 |
2.5.1 |
12.4.1 |
Release1, Release2, Release3 |
| 3.12 |
2.5.1 |
11.8.0 |
Release1 |
| 3.12 |
2.4.1 |
12.6.3 |
Release1, Release2, Release3 |
| 3.12 |
2.4.1 |
12.4.1 |
Release1, Release2, Release3 |
| 3.12 |
2.4.1 |
11.8.0 |
Release1 |
| 3.12 |
2.3.1 |
12.6.3 |
Release1, Release2 |
| 3.12 |
2.3.1 |
12.4.1 |
Release1, Release2 |
| 3.12 |
2.2.2 |
12.6.3 |
Release1, Release2 |
| 3.12 |
2.2.2 |
12.4.1 |
Release1, Release2 |
| 3.12 |
2.1.2 |
12.6.3 |
Release1 |
| 3.12 |
2.1.2 |
12.4.1 |
Release1 |
| 3.12 |
2.0.1 |
12.6.3 |
Release1 |
| 3.12 |
2.0.1 |
12.4.1 |
Release1 |
| 3.11 |
2.7.0 |
12.6.3 |
Release1 |
| 3.11 |
2.7.0 |
12.4.1 |
Release1 |
| 3.11 |
2.7.0 |
11.8.0 |
Release1 |
| 3.11 |
2.6.0 |
12.6.3 |
Release1, Release2, Release3 |
| 3.11 |
2.6.0 |
12.4.1 |
Release1, Release2, Release3 |
| 3.11 |
2.6.0 |
11.8.0 |
Release1 |
| 3.11 |
2.5.1 |
12.6.3 |
Release1, Release2, Release3 |
| 3.11 |
2.5.1 |
12.4.1 |
Release1, Release2, Release3 |
| 3.11 |
2.5.1 |
11.8.0 |
Release1 |
| 3.11 |
2.4.1 |
12.6.3 |
Release1, Release2, Release3 |
| 3.11 |
2.4.1 |
12.4.1 |
Release1, Release2, Release3 |
| 3.11 |
2.4.1 |
11.8.0 |
Release1 |
| 3.11 |
2.3.1 |
12.6.3 |
Release1, Release2 |
| 3.11 |
2.3.1 |
12.4.1 |
Release1, Release2 |
| 3.11 |
2.2.2 |
12.6.3 |
Release1, Release2 |
| 3.11 |
2.2.2 |
12.4.1 |
Release1, Release2 |
| 3.11 |
2.1.2 |
12.6.3 |
Release1 |
| 3.11 |
2.1.2 |
12.4.1 |
Release1 |
| 3.11 |
2.0.1 |
12.6.3 |
Release1 |
| 3.11 |
2.0.1 |
12.4.1 |
Release1 |
| 3.10 |
2.7.0 |
12.6.3 |
Release1 |
| 3.10 |
2.7.0 |
12.4.1 |
Release1 |
| 3.10 |
2.7.0 |
11.8.0 |
Release1 |
| 3.10 |
2.6.0 |
12.6.3 |
Release1, Release2, Release3 |
| 3.10 |
2.6.0 |
12.4.1 |
Release1, Release2, Release3 |
| 3.10 |
2.6.0 |
11.8.0 |
Release1 |
| 3.10 |
2.5.1 |
12.6.3 |
Release1, Release2, Release3 |
| 3.10 |
2.5.1 |
12.4.1 |
Release1, Release2, Release3 |
| 3.10 |
2.5.1 |
11.8.0 |
Release1 |
| 3.10 |
2.4.1 |
12.6.3 |
Release1, Release2, Release3 |
| 3.10 |
2.4.1 |
12.4.1 |
Release1, Release2, Release3 |
| 3.10 |
2.4.1 |
11.8.0 |
Release1 |
| 3.10 |
2.3.1 |
12.6.3 |
Release1, Release2 |
| 3.10 |
2.3.1 |
12.4.1 |
Release1, Release2 |
| 3.10 |
2.2.2 |
12.6.3 |
Release1, Release2 |
| 3.10 |
2.2.2 |
12.4.1 |
Release1, Release2 |
| 3.10 |
2.1.2 |
12.6.3 |
Release1 |
| 3.10 |
2.1.2 |
12.4.1 |
Release1 |
| 3.10 |
2.0.1 |
12.6.3 |
Release1 |
| 3.10 |
2.0.1 |
12.4.1 |
Release1 |
Flash-Attention 2.7.3
Packages for Flash-Attention 2.7.3
Flash-Attention 2.7.2.post1
Packages for Flash-Attention 2.7.2.post1
Flash-Attention 2.7.0.post2
Packages for Flash-Attention 2.7.0.post2
Flash-Attention 2.6.3
Packages for Flash-Attention 2.6.3
| Python |
PyTorch |
CUDA |
package |
| 3.13 |
2.9 |
13.0 |
Release1 |
| 3.13 |
2.9 |
12.8 |
Release1 |
| 3.13 |
2.9 |
12.6 |
Release1 |
| 3.12 |
2.9 |
13.0 |
Release1 |
| 3.12 |
2.9 |
12.8 |
Release1 |
| 3.12 |
2.9 |
12.6 |
Release1 |
| 3.12 |
2.8.0 |
12.9.1 |
Release1 |
| 3.12 |
2.8.0 |
12.8.1 |
Release1 |
| 3.12 |
2.8.0 |
12.4.1 |
Release1 |
| 3.12 |
2.8.0.dev20250523 |
12.8.1 |
Release1, Release2 |
| 3.12 |
2.7.1 |
12.9.1 |
Release1 |
| 3.12 |
2.7.1 |
12.8.1 |
Release1, Release2 |
| 3.12 |
2.7.1 |
12.4.1 |
Release1 |
| 3.12 |
2.7.0 |
12.8.1 |
Release1, Release2 |
| 3.12 |
2.7.0 |
12.6.3 |
Release1 |
| 3.12 |
2.7.0 |
12.4.1 |
Release1 |
| 3.12 |
2.7.0 |
11.8.0 |
Release1 |
| 3.12 |
2.6.0 |
12.9.1 |
Release1 |
| 3.12 |
2.6.0 |
12.8.1 |
Release1 |
| 3.12 |
2.6.0 |
12.6.3 |
Release1, Release2, Release3 |
| 3.12 |
2.6.0 |
12.4.1 |
Release1, Release2, Release3, Release4 |
| 3.12 |
2.6.0 |
11.8.0 |
Release1 |
| 3.12 |
2.5.1 |
12.9.1 |
Release1 |
| 3.12 |
2.5.1 |
12.8.1 |
Release1 |
| 3.12 |
2.5.1 |
12.6.3 |
Release1, Release2, Release3 |
| 3.12 |
2.5.1 |
12.4.1 |
Release1, Release2, Release3, Release4, Release5 |
| 3.12 |
2.5.1 |
12.1.1 |
Release1 |
| 3.12 |
2.5.1 |
11.8.0 |
Release1, Release2 |
| 3.12 |
2.4.1 |
12.6.3 |
Release1, Release2, Release3 |
| 3.12 |
2.4.1 |
12.4.1 |
Release1, Release2, Release3, Release4 |
| 3.12 |
2.4.1 |
12.1.1 |
Release1 |
| 3.12 |
2.4.1 |
11.8.0 |
Release1, Release2 |
| 3.12 |
2.3.1 |
12.6.3 |
Release1, Release2 |
| 3.12 |
2.3.1 |
12.4.1 |
Release1, Release2, Release3 |
| 3.12 |
2.3.1 |
12.1.1 |
Release1 |
| 3.12 |
2.3.1 |
11.8.0 |
Release1 |
| 3.12 |
2.2.2 |
12.6.3 |
Release1, Release2 |
| 3.12 |
2.2.2 |
12.4.1 |
Release1, Release2, Release3 |
| 3.12 |
2.2.2 |
12.1.1 |
Release1 |
| 3.12 |
2.2.2 |
11.8.0 |
Release1 |
| 3.12 |
2.1.2 |
12.6.3 |
Release1 |
| 3.12 |
2.1.2 |
12.4.1 |
Release1, Release2 |
| 3.12 |
2.1.2 |
12.1.1 |
Release1 |
| 3.12 |
2.1.2 |
11.8.0 |
Release1 |
| 3.12 |
2.0.1 |
12.6.3 |
Release1 |
| 3.12 |
2.0.1 |
12.4.1 |
Release1, Release2 |
| 3.12 |
2.0.1 |
12.1.1 |
Release1 |
| 3.12 |
2.0.1 |
11.8.0 |
Release1 |
| 3.11 |
2.9 |
13.0 |
Release1 |
| 3.11 |
2.9 |
12.8 |
Release1 |
| 3.11 |
2.9 |
12.6 |
Release1 |
| 3.11 |
2.8.0 |
12.9.1 |
Release1 |
| 3.11 |
2.8.0 |
12.8.1 |
Release1 |
| 3.11 |
2.8.0 |
12.4.1 |
Release1 |
| 3.11 |
2.8.0.dev20250523 |
12.8.1 |
Release1, Release2 |
| 3.11 |
2.7.1 |
12.9.1 |
Release1 |
| 3.11 |
2.7.1 |
12.8.1 |
Release1, Release2 |
| 3.11 |
2.7.1 |
12.4.1 |
Release1 |
| 3.11 |
2.7.0 |
12.8.1 |
Release1, Release2 |
| 3.11 |
2.7.0 |
12.6.3 |
Release1 |
| 3.11 |
2.7.0 |
12.4.1 |
Release1 |
| 3.11 |
2.7.0 |
11.8.0 |
Release1 |
| 3.11 |
2.6.0 |
12.9.1 |
Release1 |
| 3.11 |
2.6.0 |
12.8.1 |
Release1 |
| 3.11 |
2.6.0 |
12.6.3 |
Release1, Release2, Release3 |
| 3.11 |
2.6.0 |
12.4.1 |
Release1, Release2, Release3, Release4 |
| 3.11 |
2.6.0 |
11.8.0 |
Release1 |
| 3.11 |
2.5.1 |
12.9.1 |
Release1 |
| 3.11 |
2.5.1 |
12.8.1 |
Release1 |
| 3.11 |
2.5.1 |
12.6.3 |
Release1, Release2, Release3 |
| 3.11 |
2.5.1 |
12.4.1 |
Release1, Release2, Release3, Release4, Release5 |
| 3.11 |
2.5.1 |
12.1.1 |
Release1 |
| 3.11 |
2.5.1 |
11.8.0 |
Release1, Release2 |
| 3.11 |
2.4.1 |
12.6.3 |
Release1, Release2, Release3 |
| 3.11 |
2.4.1 |
12.4.1 |
Release1, Release2, Release3, Release4 |
| 3.11 |
2.4.1 |
12.1.1 |
Release1 |
| 3.11 |
2.4.1 |
11.8.0 |
Release1, Release2 |
| 3.11 |
2.3.1 |
12.6.3 |
Release1, Release2 |
| 3.11 |
2.3.1 |
12.4.1 |
Release1, Release2, Release3 |
| 3.11 |
2.3.1 |
12.1.1 |
Release1 |
| 3.11 |
2.3.1 |
11.8.0 |
Release1 |
| 3.11 |
2.2.2 |
12.6.3 |
Release1, Release2 |
| 3.11 |
2.2.2 |
12.4.1 |
Release1, Release2, Release3 |
| 3.11 |
2.2.2 |
12.1.1 |
Release1 |
| 3.11 |
2.2.2 |
11.8.0 |
Release1 |
| 3.11 |
2.1.2 |
12.6.3 |
Release1 |
| 3.11 |
2.1.2 |
12.4.1 |
Release1, Release2 |
| 3.11 |
2.1.2 |
12.1.1 |
Release1 |
| 3.11 |
2.1.2 |
11.8.0 |
Release1 |
| 3.11 |
2.0.1 |
12.6.3 |
Release1 |
| 3.11 |
2.0.1 |
12.4.1 |
Release1, Release2 |
| 3.11 |
2.0.1 |
12.1.1 |
Release1 |
| 3.11 |
2.0.1 |
11.8.0 |
Release1 |
| 3.10 |
2.9 |
13.0 |
Release1 |
| 3.10 |
2.9 |
12.8 |
Release1 |
| 3.10 |
2.9 |
12.6 |
Release1 |
| 3.10 |
2.8.0 |
12.9.1 |
Release1 |
| 3.10 |
2.8.0 |
12.8.1 |
Release1 |
| 3.10 |
2.8.0 |
12.4.1 |
Release1 |
| 3.10 |
2.8.0.dev20250523 |
12.8.1 |
Release1, Release2 |
| 3.10 |
2.7.1 |
12.9.1 |
Release1 |
| 3.10 |
2.7.1 |
12.8.1 |
Release1, Release2 |
| 3.10 |
2.7.1 |
12.4.1 |
Release1 |
| 3.10 |
2.7.0 |
12.8.1 |
Release1, Release2 |
| 3.10 |
2.7.0 |
12.6.3 |
Release1 |
| 3.10 |
2.7.0 |
12.4.1 |
Release1 |
| 3.10 |
2.7.0 |
11.8.0 |
Release1 |
| 3.10 |
2.6.0 |
12.9.1 |
Release1 |
| 3.10 |
2.6.0 |
12.8.1 |
Release1 |
| 3.10 |
2.6.0 |
12.6.3 |
Release1, Release2, Release3 |
| 3.10 |
2.6.0 |
12.4.1 |
Release1, Release2, Release3, Release4 |
| 3.10 |
2.6.0 |
11.8.0 |
Release1 |
| 3.10 |
2.5.1 |
12.9.1 |
Release1 |
| 3.10 |
2.5.1 |
12.8.1 |
Release1 |
| 3.10 |
2.5.1 |
12.6.3 |
Release1, Release2, Release3 |
| 3.10 |
2.5.1 |
12.4.1 |
Release1, Release2, Release3, Release4, Release5 |
| 3.10 |
2.5.1 |
12.1.1 |
Release1 |
| 3.10 |
2.5.1 |
11.8.0 |
Release1, Release2 |
| 3.10 |
2.4.1 |
12.6.3 |
Release1, Release2, Release3 |
| 3.10 |
2.4.1 |
12.4.1 |
Release1, Release2, Release3, Release4 |
| 3.10 |
2.4.1 |
12.1.1 |
Release1 |
| 3.10 |
2.4.1 |
11.8.0 |
Release1, Release2 |
| 3.10 |
2.3.1 |
12.6.3 |
Release1, Release2 |
| 3.10 |
2.3.1 |
12.4.1 |
Release1, Release2, Release3 |
| 3.10 |
2.3.1 |
12.1.1 |
Release1 |
| 3.10 |
2.3.1 |
11.8.0 |
Release1 |
| 3.10 |
2.2.2 |
12.6.3 |
Release1, Release2 |
| 3.10 |
2.2.2 |
12.4.1 |
Release1, Release2, Release3 |
| 3.10 |
2.2.2 |
12.1.1 |
Release1 |
| 3.10 |
2.2.2 |
11.8.0 |
Release1 |
| 3.10 |
2.1.2 |
12.6.3 |
Release1 |
| 3.10 |
2.1.2 |
12.4.1 |
Release1, Release2 |
| 3.10 |
2.1.2 |
12.1.1 |
Release1 |
| 3.10 |
2.1.2 |
11.8.0 |
Release1 |
| 3.10 |
2.0.1 |
12.6.3 |
Release1 |
| 3.10 |
2.0.1 |
12.4.1 |
Release1, Release2 |
| 3.10 |
2.0.1 |
12.1.1 |
Release1 |
| 3.10 |
2.0.1 |
11.8.0 |
Release1 |
| 3.9 |
2.8 |
12.6 |
Release1 |
| 3.9 |
2.8 |
12.4 |
Release1 |
| 3.9 |
2.7 |
12.6 |
Release1 |
| 3.9 |
2.7 |
12.4 |
Release1 |
| 3.9 |
2.6 |
12.6 |
Release1 |
| 3.9 |
2.6 |
12.4 |
Release1 |
| 3.9 |
2.5 |
12.6 |
Release1 |
| 3.9 |
2.5 |
12.4 |
Release1 |
Flash-Attention 2.5.9
Packages for Flash-Attention 2.5.9
| Python |
PyTorch |
CUDA |
package |
| 3.12 |
2.8.0.dev20250523 |
12.8.1 |
Release1, Release2 |
| 3.12 |
2.7.1 |
12.8.1 |
Release1 |
| 3.12 |
2.7.0 |
12.8.1 |
Release1, Release2 |
| 3.12 |
2.7.0 |
12.6.3 |
Release1 |
| 3.12 |
2.7.0 |
12.4.1 |
Release1 |
| 3.12 |
2.7.0 |
11.8.0 |
Release1 |
| 3.12 |
2.6.0 |
12.6.3 |
Release1, Release2 |
| 3.12 |
2.6.0 |
12.4.1 |
Release1, Release2 |
| 3.12 |
2.6.0 |
11.8.0 |
Release1 |
| 3.12 |
2.5.1 |
12.6.3 |
Release1, Release2 |
| 3.12 |
2.5.1 |
12.4.1 |
Release1, Release2 |
| 3.12 |
2.5.1 |
11.8.0 |
Release1 |
| 3.12 |
2.4.1 |
12.6.3 |
Release1, Release2 |
| 3.12 |
2.4.1 |
12.4.1 |
Release1, Release2 |
| 3.12 |
2.4.1 |
11.8.0 |
Release1 |
| 3.12 |
2.3.1 |
12.6.3 |
Release1 |
| 3.12 |
2.3.1 |
12.4.1 |
Release1 |
| 3.12 |
2.2.2 |
12.6.3 |
Release1 |
| 3.12 |
2.2.2 |
12.4.1 |
Release1 |
| 3.11 |
2.8.0.dev20250523 |
12.8.1 |
Release1, Release2 |
| 3.11 |
2.7.1 |
12.8.1 |
Release1 |
| 3.11 |
2.7.0 |
12.8.1 |
Release1, Release2 |
| 3.11 |
2.7.0 |
12.6.3 |
Release1 |
| 3.11 |
2.7.0 |
12.4.1 |
Release1 |
| 3.11 |
2.7.0 |
11.8.0 |
Release1 |
| 3.11 |
2.6.0 |
12.6.3 |
Release1, Release2 |
| 3.11 |
2.6.0 |
12.4.1 |
Release1, Release2 |
| 3.11 |
2.6.0 |
11.8.0 |
Release1 |
| 3.11 |
2.5.1 |
12.6.3 |
Release1, Release2 |
| 3.11 |
2.5.1 |
12.4.1 |
Release1, Release2 |
| 3.11 |
2.5.1 |
11.8.0 |
Release1 |
| 3.11 |
2.4.1 |
12.6.3 |
Release1, Release2 |
| 3.11 |
2.4.1 |
12.4.1 |
Release1, Release2 |
| 3.11 |
2.4.1 |
11.8.0 |
Release1 |
| 3.11 |
2.3.1 |
12.6.3 |
Release1 |
| 3.11 |
2.3.1 |
12.4.1 |
Release1 |
| 3.11 |
2.2.2 |
12.6.3 |
Release1 |
| 3.11 |
2.2.2 |
12.4.1 |
Release1 |
| 3.10 |
2.8.0.dev20250523 |
12.8.1 |
Release1, Release2 |
| 3.10 |
2.7.1 |
12.8.1 |
Release1 |
| 3.10 |
2.7.0 |
12.8.1 |
Release1, Release2 |
| 3.10 |
2.7.0 |
12.6.3 |
Release1 |
| 3.10 |
2.7.0 |
12.4.1 |
Release1 |
| 3.10 |
2.7.0 |
11.8.0 |
Release1 |
| 3.10 |
2.6.0 |
12.6.3 |
Release1, Release2 |
| 3.10 |
2.6.0 |
12.4.1 |
Release1, Release2 |
| 3.10 |
2.6.0 |
11.8.0 |
Release1 |
| 3.10 |
2.5.1 |
12.6.3 |
Release1, Release2 |
| 3.10 |
2.5.1 |
12.4.1 |
Release1, Release2 |
| 3.10 |
2.5.1 |
11.8.0 |
Release1 |
| 3.10 |
2.4.1 |
12.6.3 |
Release1, Release2 |
| 3.10 |
2.4.1 |
12.4.1 |
Release1, Release2 |
| 3.10 |
2.4.1 |
11.8.0 |
Release1 |
| 3.10 |
2.3.1 |
12.6.3 |
Release1 |
| 3.10 |
2.3.1 |
12.4.1 |
Release1 |
| 3.10 |
2.2.2 |
12.6.3 |
Release1 |
| 3.10 |
2.2.2 |
12.4.1 |
Release1 |
Flash-Attention 2.5.6
Packages for Flash-Attention 2.5.6
Flash-Attention 2.4.3
Packages for Flash-Attention 2.4.3
| Python |
PyTorch |
CUDA |
package |
| 3.12 |
2.8.0.dev20250523 |
12.8.1 |
Release1, Release2 |
| 3.12 |
2.7.1 |
12.8.1 |
Release1 |
| 3.12 |
2.7.0 |
12.8.1 |
Release1, Release2 |
| 3.12 |
2.7.0 |
12.6.3 |
Release1 |
| 3.12 |
2.7.0 |
12.4.1 |
Release1 |
| 3.12 |
2.7.0 |
11.8.0 |
Release1 |
| 3.12 |
2.6.0 |
12.6.3 |
Release1, Release2 |
| 3.12 |
2.6.0 |
12.4.1 |
Release1, Release2 |
| 3.12 |
2.6.0 |
11.8.0 |
Release1 |
| 3.12 |
2.5.1 |
12.6.3 |
Release1, Release2 |
| 3.12 |
2.5.1 |
12.4.1 |
Release1, Release2, Release3 |
| 3.12 |
2.5.1 |
12.1.1 |
Release1 |
| 3.12 |
2.5.1 |
11.8.0 |
Release1, Release2 |
| 3.12 |
2.4.1 |
12.6.3 |
Release1, Release2 |
| 3.12 |
2.4.1 |
12.4.1 |
Release1, Release2, Release3 |
| 3.12 |
2.4.1 |
12.1.1 |
Release1 |
| 3.12 |
2.4.1 |
11.8.0 |
Release1, Release2 |
| 3.12 |
2.3.1 |
12.6.3 |
Release1 |
| 3.12 |
2.3.1 |
12.4.1 |
Release1, Release2 |
| 3.12 |
2.3.1 |
12.1.1 |
Release1 |
| 3.12 |
2.3.1 |
11.8.0 |
Release1 |
| 3.12 |
2.2.2 |
12.6.3 |
Release1 |
| 3.12 |
2.2.2 |
12.4.1 |
Release1, Release2 |
| 3.12 |
2.2.2 |
12.1.1 |
Release1 |
| 3.12 |
2.2.2 |
11.8.0 |
Release1 |
| 3.12 |
2.1.2 |
12.4.1 |
Release1 |
| 3.12 |
2.1.2 |
12.1.1 |
Release1 |
| 3.12 |
2.1.2 |
11.8.0 |
Release1 |
| 3.12 |
2.0.1 |
12.4.1 |
Release1 |
| 3.12 |
2.0.1 |
12.1.1 |
Release1 |
| 3.12 |
2.0.1 |
11.8.0 |
Release1 |
| 3.11 |
2.8.0.dev20250523 |
12.8.1 |
Release1, Release2 |
| 3.11 |
2.7.1 |
12.8.1 |
Release1 |
| 3.11 |
2.7.0 |
12.8.1 |
Release1, Release2 |
| 3.11 |
2.7.0 |
12.6.3 |
Release1 |
| 3.11 |
2.7.0 |
12.4.1 |
Release1 |
| 3.11 |
2.7.0 |
11.8.0 |
Release1 |
| 3.11 |
2.6.0 |
12.6.3 |
Release1, Release2 |
| 3.11 |
2.6.0 |
12.4.1 |
Release1, Release2 |
| 3.11 |
2.6.0 |
11.8.0 |
Release1 |
| 3.11 |
2.5.1 |
12.6.3 |
Release1, Release2 |
| 3.11 |
2.5.1 |
12.4.1 |
Release1, Release2, Release3 |
| 3.11 |
2.5.1 |
12.1.1 |
Release1 |
| 3.11 |
2.5.1 |
11.8.0 |
Release1, Release2 |
| 3.11 |
2.4.1 |
12.6.3 |
Release1, Release2 |
| 3.11 |
2.4.1 |
12.4.1 |
Release1, Release2, Release3 |
| 3.11 |
2.4.1 |
12.1.1 |
Release1 |
| 3.11 |
2.4.1 |
11.8.0 |
Release1, Release2 |
| 3.11 |
2.3.1 |
12.6.3 |
Release1 |
| 3.11 |
2.3.1 |
12.4.1 |
Release1, Release2 |
| 3.11 |
2.3.1 |
12.1.1 |
Release1 |
| 3.11 |
2.3.1 |
11.8.0 |
Release1 |
| 3.11 |
2.2.2 |
12.6.3 |
Release1 |
| 3.11 |
2.2.2 |
12.4.1 |
Release1, Release2 |
| 3.11 |
2.2.2 |
12.1.1 |
Release1 |
| 3.11 |
2.2.2 |
11.8.0 |
Release1 |
| 3.11 |
2.1.2 |
12.4.1 |
Release1 |
| 3.11 |
2.1.2 |
12.1.1 |
Release1 |
| 3.11 |
2.1.2 |
11.8.0 |
Release1 |
| 3.11 |
2.0.1 |
12.4.1 |
Release1 |
| 3.11 |
2.0.1 |
12.1.1 |
Release1 |
| 3.11 |
2.0.1 |
11.8.0 |
Release1 |
| 3.10 |
2.8.0.dev20250523 |
12.8.1 |
Release1, Release2 |
| 3.10 |
2.7.1 |
12.8.1 |
Release1 |
| 3.10 |
2.7.0 |
12.8.1 |
Release1, Release2 |
| 3.10 |
2.7.0 |
12.6.3 |
Release1 |
| 3.10 |
2.7.0 |
12.4.1 |
Release1 |
| 3.10 |
2.7.0 |
11.8.0 |
Release1 |
| 3.10 |
2.6.0 |
12.6.3 |
Release1, Release2 |
| 3.10 |
2.6.0 |
12.4.1 |
Release1, Release2 |
| 3.10 |
2.6.0 |
11.8.0 |
Release1 |
| 3.10 |
2.5.1 |
12.6.3 |
Release1, Release2 |
| 3.10 |
2.5.1 |
12.4.1 |
Release1, Release2, Release3 |
| 3.10 |
2.5.1 |
12.1.1 |
Release1 |
| 3.10 |
2.5.1 |
11.8.0 |
Release1, Release2 |
| 3.10 |
2.4.1 |
12.6.3 |
Release1, Release2 |
| 3.10 |
2.4.1 |
12.4.1 |
Release1, Release2, Release3 |
| 3.10 |
2.4.1 |
12.1.1 |
Release1 |
| 3.10 |
2.4.1 |
11.8.0 |
Release1, Release2 |
| 3.10 |
2.3.1 |
12.6.3 |
Release1 |
| 3.10 |
2.3.1 |
12.4.1 |
Release1, Release2 |
| 3.10 |
2.3.1 |
12.1.1 |
Release1 |
| 3.10 |
2.3.1 |
11.8.0 |
Release1 |
| 3.10 |
2.2.2 |
12.6.3 |
Release1 |
| 3.10 |
2.2.2 |
12.4.1 |
Release1, Release2 |
| 3.10 |
2.2.2 |
12.1.1 |
Release1 |
| 3.10 |
2.2.2 |
11.8.0 |
Release1 |
| 3.10 |
2.1.2 |
12.4.1 |
Release1 |
| 3.10 |
2.1.2 |
12.1.1 |
Release1 |
| 3.10 |
2.1.2 |
11.8.0 |
Release1 |
| 3.10 |
2.0.1 |
12.4.1 |
Release1 |
| 3.10 |
2.0.1 |
12.1.1 |
Release1 |
| 3.10 |
2.0.1 |
11.8.0 |
Release1 |
Windows x86_64
Flash-Attention 2.8.3
Packages for Flash-Attention 2.8.3
Flash-Attention 2.8.2
Packages for Flash-Attention 2.8.2
Flash-Attention 2.7.4
Packages for Flash-Attention 2.7.4
Flash-Attention 2.6.3
Packages for Flash-Attention 2.6.3
Flash-Attention 2.5.9
Packages for Flash-Attention 2.5.9
History
v0.4.18
Release
Linux x86_64
| Flash-Attention |
Python |
PyTorch |
CUDA |
| 2.6.3, 2.8.3 |
3.10, 3.11, 3.12, 3.13 |
2.9 |
13.0 |
v0.4.17
Release
Linux x86_64
| Flash-Attention |
Python |
PyTorch |
CUDA |
| 2.6.3, 2.8.3 |
3.10, 3.11, 3.12, 3.13 |
2.9 |
12.6, 12.8 |
v0.4.16
Release
Linux x86_64
| Flash-Attention |
Python |
PyTorch |
CUDA |
| 2.6.3, 2.8.3 |
3.9 |
2.5, 2.6, 2.7, 2.8 |
12.4, 12.6 |
v0.4.15
Release
Linux x86_64
| Flash-Attention |
Python |
PyTorch |
CUDA |
| 2.8.3 |
3.11, 3.12, 3.13 |
2.9 |
12.6, 12.8 |
Windows x86_64
| Flash-Attention |
Python |
PyTorch |
CUDA |
| 2.8.3 |
3.11, 3.12, 3.13 |
2.9 |
12.6 |
v0.4.12
Release
Linux x86_64
| Flash-Attention |
Python |
PyTorch |
CUDA |
| 2.8.3 |
3.13 |
2.6, 2.7, 2.8 |
12.4, 12.6, 12.8, 12.9 |
Windows x86_64
| Flash-Attention |
Python |
PyTorch |
CUDA |
| 2.8.2 |
3.13 |
2.6, 2.7, 2.8 |
12.4, 12.6 |
v0.4.11
Release
Linux x86_64
| Flash-Attention |
Python |
PyTorch |
CUDA |
| 2.8.3 |
3.10, 3.11, 3.12 |
2.5, 2.6, 2.7, 2.8 |
12.4, 12.6, 12.8, 12.9 |
v0.4.10
Release
Windows x86_64
| Flash-Attention |
Python |
PyTorch |
CUDA |
| 2.7.4, 2.8.2 |
3.10, 3.11, 3.12 |
2.7, 2.8 |
12.8 |
v0.4.9
Release
Windows x86_64
| Flash-Attention |
Python |
PyTorch |
CUDA |
| 2.7.4 |
3.11 |
2.7 |
12.8 |
v0.3.18
Release
Linux x86_64
| Flash-Attention |
Python |
PyTorch |
CUDA |
| 2.7.4 |
3.10, 3.11, 3.12 |
2.5, 2.6, 2.7, 2.8 |
12.4, 12.8, 12.9 |
v0.3.14
Release
Linux x86_64
| Flash-Attention |
Python |
PyTorch |
CUDA |
| 2.6.3, 2.8.2 |
3.10, 3.11, 3.12 |
2.5.1, 2.6.0, 2.7.1, 2.8.0 |
12.4.1, 12.8.1, 12.9.1 |
v0.3.13
Release
Linux x86_64
| Flash-Attention |
Python |
PyTorch |
CUDA |
| 2.8.1 |
3.10, 3.11, 3.12 |
2.4.1, 2.5.1, 2.6.0, 2.7.1 |
12.8.1 |
v0.3.12
Release
Linux x86_64
| Flash-Attention |
Python |
PyTorch |
CUDA |
| 2.8.0 |
3.10, 3.11, 3.12 |
2.4.1, 2.5.1, 2.6.0, 2.7.1 |
12.4.1, 12.8.1 |
v0.3.10
Release
Linux x86_64
| Flash-Attention |
Python |
PyTorch |
CUDA |
| 2.7.4 |
3.10, 3.11, 3.12 |
2.7.1 |
12.8.1 |
v0.3.9
Release
Linux x86_64
| Flash-Attention |
Python |
PyTorch |
CUDA |
| 2.4.3, 2.5.9, 2.6.3 |
3.10, 3.11, 3.12 |
2.7.1 |
12.8.1 |
Windows x86_64
| Flash-Attention |
Python |
PyTorch |
CUDA |
| 2.5.9, 2.6.3, 2.7.4 |
3.10, 3.11, 3.12 |
2.4.1, 2.5.1, 2.6.0 |
12.4.1 |
v0.3.1
Release
Windows x86_64
| Flash-Attention |
Python |
PyTorch |
CUDA |
| 2.6.3 |
3.11 |
2.6.0 |
12.6.3 |
From this version, Wheels for Windows are released.
However, we are waiting for a report on how it works because we have not tested it enough.
v0.2.1
Release
| Flash-Attention |
Python |
PyTorch |
CUDA |
| 2.4.3, 2.5.9, 2.6.3, 2.7.4 |
3.10, 3.11, 3.12 |
2.8.0.dev20250523 |
12.8.1 |
v0.2.0
Release
| Flash-Attention |
Python |
PyTorch |
CUDA |
| 2.4.3, 2.5.9, 2.6.3 |
3.10, 3.11, 3.12 |
2.8.0.dev20250523 |
12.8.1 |
v0.1.0
Release
| Flash-Attention |
Python |
PyTorch |
CUDA |
| 2.4.3, 2.5.9, 2.6.3, 2.7.4 |
3.10, 3.11, 3.12 |
2.7.0 |
12.8.1 |
v2.7.4 and v2.7.4.post1 are the same version.
From this release, self-hosted runners are used for building some wheels.
v0.0.9
Release
| Flash-Attention |
Python |
PyTorch |
CUDA |
| 2.4.3, 2.5.9, 2.6.3 |
3.10, 3.11, 3.12 |
2.7.0 |
12.8.1 |
v0.0.8
Release
| Flash-Attention |
Python |
PyTorch |
CUDA |
| 2.4.3, 2.5.9, 2.6.3, 2.7.4.post1 |
3.10, 3.11, 3.12 |
2.4.1, 2.5.1, 2.6.0, 2.7.0 |
11.8.0, 12.4.1, 12.6.3 |
v0.0.7
Skip for experimental reasons.
v0.0.6
Release
| Flash-Attention |
Python |
PyTorch |
CUDA |
| 2.4.3, 2.5.9, 2.6.3, 2.7.4.post1 |
3.10, 3.11, 3.12 |
2.2.2, 2.3.1, 2.4.1, 2.5.1, 2.6.0 |
12.4.1, 12.6.3 |
v0.0.5
Release
| Flash-Attention |
Python |
PyTorch |
CUDA |
| 2.6.3, 2.7.4.post1 |
3.10, 3.11, 3.12 |
2.0.1, 2.1.2, 2.2.2, 2.3.1, 2.4.1, 2.5.1, 2.6.0 |
12.4.1, 12.6.3 |
v0.0.4
Release
| Flash-Attention |
Python |
PyTorch |
CUDA |
| 2.7.3 |
3.10, 3.11, 3.12 |
2.0.1, 2.1.2, 2.2.2, 2.3.1, 2.4.1, 2.5.1 |
11.8.0, 12.1.1, 12.4.1 |
v0.0.3
Release
| Flash-Attention |
Python |
PyTorch |
CUDA |
| 2.7.2.post1 |
3.10, 3.11, 3.12 |
2.0.1, 2.1.2, 2.2.2, 2.3.1, 2.4.1, 2.5.1 |
11.8.0, 12.1.1, 12.4.1 |
v0.0.2
Release
| Flash-Attention |
Python |
PyTorch |
CUDA |
| 2.4.3, 2.5.6, 2.6.3, 2.7.0.post2 |
3.10, 3.11, 3.12 |
2.0.1, 2.1.2, 2.2.2, 2.3.1, 2.4.1, 2.5.1 |
11.8.0, 12.1.1, 12.4.1 |
v0.0.1
Release
| flash-attention |
Python |
PyTorch |
CUDA |
| 1.0.9, 2.4.3, 2.5.6, 2.5.9, 2.6.3 |
3.10, 3.11, 3.12 |
2.0.1, 2.1.2, 2.2.2, 2.3.1, 2.4.1, 2.5.0 |
11.8.0, 12.1.1, 12.4.1 |
v0.0.0
Release
| flash-attention |
Python |
PyTorch |
CUDA |
| 2.4.3, 2.5.6, 2.5.9, 2.6.3 |
3.11, 3.12 |
2.0.1, 2.1.2, 2.2.2, 2.3.1, 2.4.1, 2.5.0 |
11.8.0, 12.1.1, 12.4.1 |
Original Repository
repo