X · @teortaxesTex
· X / Twitter
probably the best implementation of linear attention to date
probably the best implementation of linear attention to dateKimi.ai: We've open-sourced FlashKDA, our high-performance CUTLASS-based implementation of Kimi Delta Attention kernels.It delivers 1.72×–2.22× prefill speedup over the flash-linear-attention baseline on H20, and works as a drop-in backend for flash-linear-att