tspeterkim/flash-attention-minimal
Flash Attention in ~100 lines of CUDA (forward pass only)
CudaApache-2.0
Issues
- 2
Does this rep surport tensorcore?
#5 opened by Rane2021 - 0
- 2
slow in for loop test
#3 opened by DefTruth - 1
Correctness parameters
#1 opened by cogumbreiro