BetaRare
thu-ml/SpargeAttn
[ICML2025] SpargeAttention: A training-free sparse attention that accelerates any model inference.
- Owner
- thu-ml (organization)
- Stars
- Stars: 1,266
- Forks
- Forks: 108
- Language
- Cuda
- License
- Apache-2.0
- Created
- 2025
- Topics
- ai-infraattentioninference-accelerationllmmlsysquantizationsparse-attentionvision-transformersageattentionvideo-generation
- Tier
- Beta (700 to 1,999 ★) What this tier means
More repos by thu-ml
All 27 repos by thu-ml- thu-ml/tianshouStars: 11,048PythonStable
- thu-ml/SageAttentionStars: 3,988CudaRC
- thu-ml/TurboDiffusionStars: 3,857PythonRC
- thu-ml/zhusuanStars: 2,218PythonRC
- thu-ml/RoboticsDiffusionTransformerStars: 1,815PythonBeta
- thu-ml/prolificdreamerStars: 1,563PythonBeta
Nearby in Cuda
All Cuda repos- siboehm/SGEMM_CUDAStars: 1,355CudaBeta
- MoonshotAI/FlashKDAStars: 1,295CudaBeta
- PacktPublishing/Learn-CUDA-ProgrammingStars: 1,268CudaBeta
- tspeterkim/flash-attention-minimalStars: 1,203CudaBeta
- antonmks/AlenkaStars: 1,166CudaBeta
- Tongkaio/CUDA_Kernel_SamplesStars: 1,143CudaBeta