HKUSTDial/flash-sparse-attention

flash-sparse-attention

✅ Recently maintained

Trainable fast and memory-efficient sparse attention

762Stars
54Forks
+37-day stars

Python · BSD-3-Clause · Last push 2026-10-29

Official repository & installation ↗

GitValue · Value evidence

Unscored · Evidence coverage 0%

Provisional rubric; documentation and runtime testing are separate. Missing evidence does not imply low value.

Star history

2025-09-292026-11-03
View historical data
DateStars
2026-10-05756
2026-10-06756
2026-10-07756
2026-10-08756
2026-10-09756
2026-10-10756
2026-10-11756
2026-10-12756
2026-10-13756
2026-10-14757
2026-10-15758
2026-10-16758
2026-10-17758
2026-10-18758
2026-10-19758
2026-10-20758
2026-10-21758
2026-10-22759
2026-10-23759
2026-10-24759
2026-10-25759
2026-10-26759
2026-10-27759
2026-10-28760
2026-10-29760
2026-10-30760
2026-10-31761
2026-11-01761
2026-11-02762
2026-11-03762

Get the project

git clone https://github.com/HKUSTDial/flash-sparse-attention.git

Cloning downloads the source. For dependencies and installation, follow the project README.

Related projects & alternative candidates

Related domains do not imply feature parity. Review official documentation and licenses.