This release is a pre-release and may not be stable for production use.
FlashAttention-4 (CuTeDSL)
FlashAttention-4 is a CuTeDSL-based implementation of FlashAttention for Hopper and Blackwell GPUs.
Installation
pip install flash-attn-4
If you're on CUDA 13, install with the cu13 extra for best performance:
pip install "flash-attn-4[cu13]"
Usage
from flash_attn.cute import flash_attn_func, flash_attn_varlen_func
out = flash_attn_func(q, k, v, causal=True)
Development
git clone https://github.com/Dao-AILab/flash-attention.git
cd flash-attention
pip install -e "flash_attn/cute[dev]" # CUDA 12.x
pip install -e "flash_attn/cute[dev,cu13]" # CUDA 13.x (e.g. B200)
pytest tests/cute/
Release files for flash-attn-4 4.0.0b32
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | |
|---|---|---|---|---|
| flash_attn_4-4.0.0b32.tar.gz | 401.5 kB | With provenance | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| flash_attn_4-4.0.0b32-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 828.3 kB
Release files / flash_attn_4-4.0.0b32.tar.gz
| Download URL | flash_attn_4-4.0.0b32.tar.gz |
|---|---|
| Size | 401.5 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
8fbdca2ef29080a89a396d3be8c53b3925145c266e60cdde0c15d0eb73c8bc09
|
|
BLAKE2b-256 checksum How to use checksums |
93c583d5a216d0518bc2ca4320a9f10e7f2cd794bf543863608ca136e28112dc
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Sep 23, 2026.
Transparency logRelease files / flash_attn_4-4.0.0b32-py3-none-any.whl
| Download URL | flash_attn_4-4.0.0b32-py3-none-any.whl |
|---|---|
| Size | 426.8 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
9299da3524d579d84eb92954d3243aa095d55ebdafcf80d6a9766fd96fe8a562
|
|
BLAKE2b-256 checksum How to use checksums |
40d0b1e9ab8eefc9e67179a372c3e1a8e9f83d963c3a8e5bd4c738e55dd134c0
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Sep 23, 2026.
Transparency logSecurity for flash-attn-4 4.0.0b32
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
Full provenance
All 2 files in this release have been uploaded with provenance metadata.
Attestation for flash_attn_4-4.0.0b32.tar.gz
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Sep 23, 2026.
Transparency logA note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
