See what the GitHub community is most excited about today.
A bot automatically fetches new repositories from https://github.com/trending and sends them to the channel.
Author and maintainer: https://github.com/katursis
Post #15891
750
#cuda
FlashKDA is a fast CUDA kernel library for Kimi Delta Attention, built with CUTLASS and made for SM90+ GPUs with CUDA 12.9+ and PyTorch 2.4+. It can plug into flash-linear-attention as a backend, and on H20 it reports about 1.72× to 2.22× faster prefill than the existing baseline. This can help you run attention faster, with less deployment work, especially if you use supported hardware and the fixed K = V = 128 setup.
https://github.com/MoonshotAI/FlashKDA
FlashKDA is a fast CUDA kernel library for Kimi Delta Attention, built with CUTLASS and made for SM90+ GPUs with CUDA 12.9+ and PyTorch 2.4+. It can plug into flash-linear-attention as a backend, and on H20 it reports about 1.72× to 2.22× faster prefill than the existing baseline. This can help you run attention faster, with less deployment work, especially if you use supported hardware and the fixed K = V = 128 setup.
https://github.com/MoonshotAI/FlashKDA