thu-ml/
[ICML2025] SpargeAttention: A training-free sparse attention that accelerates any model inference.
by TSAIL group Cuda Apache-2.0 licence 108 forks Created Feb 2025 View on GitHub
- Stars
- 1,267
- 3% of thu-ml's stars
- Rank
- #53,800
- of 852,119 repos, top 7%
Among 852,119 repos
1k–1.6k stars: 23,672 repos, including SpargeAttn
top 50%top 10%top 1%top 0.1% 
Stars along the bottom, and repos in each band on a log scale
Quick stats
- Next rank
- +1 star to reach #53,758
- Next milestone
- +5,945 stars to reach the top 1%
- Popular
- One of 66,549 repos with 1k+ stars
- Yearly average
- 739 stars a year since Feb 2025