Popular repositories Loading

1. flash-attention flash-attention Public

   Fast and memory-efficient exact attention

   Python 23.6k 2.6k

2. quack quack Public

   A Quirky Assortment of CuTe Kernels

   Python 949 120

3. causal-conv1d causal-conv1d Public

   Causal depthwise conv1d in CUDA, with a PyTorch interface

   Cuda 854 178

4. sonic-moe sonic-moe Public

   Accelerating MoE with IO and Tile-aware Optimizations

   Python 661 79

5. fast-hadamard-transform fast-hadamard-transform Public

   Fast Hadamard transform in CUDA, with a PyTorch interface

   C 309 61

6. gram-newton-schulz gram-newton-schulz Public

   Fast Polar Decomposition for Muon

   Python 141 12