thu-ml/SageAttention
[ICLR2025, ICML2025, NeurIPS2025 Spotlight] Quantized Attention achieves speedup of 2-5x compared to FlashAttention, without losing end-to-end metrics across language, image, and video models.
4.0k
Stars
531
Forks
57.9
Radar Score
Rising
Momentum
Scoring Breakdown
Impact83.8
Velocity41.5
Health39.3
Growth Metrics
7-Day Stars
+8
30-Day Stars
+116
Acceleration
0.14x
Language
Cuda
Why it ranks
Steady growth — consistently collecting stars and contributors.
Top 99% globallyLast push 260d ago+8 stars / 7d+116 stars / 30d