SandAI-org/MagiAttention

MagiAttention

✅ Recently maintained

A Distributed Attention Towards Linear Scalability for Ultra-Long Context, Heterogeneous Data Training

968Stars
74Forks
+47-day stars

Python · Apache-2.0 · Last push 2026-11-03

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
2026-10-05946
2026-10-06946
2026-10-07946
2026-10-08947
2026-10-09947
2026-10-10947
2026-10-11948
2026-10-12950
2026-10-13950
2026-10-14950
2026-10-15952
2026-10-16952
2026-10-17952
2026-10-18954
2026-10-19956
2026-10-20958
2026-10-21958
2026-10-22960
2026-10-23961
2026-10-24963
2026-10-25963
2026-10-26964
2026-10-27964
2026-10-28965
2026-10-29965
2026-10-30967
2026-10-31967
2026-11-01967
2026-11-02968
2026-11-03968

Get the project

git clone https://github.com/SandAI-org/MagiAttention.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.