Moonshot AI open-sources FlashKDA, a CUTLASS-based Kimi Delta Attention kernel
Original titleWe've open-sourced FlashKDA, our high-performance CUTLASS-based implementation of Kimi Delta Attention kernels.
AISummary
Moonshot AI has open-sourced FlashKDA, a high-performance CUTLASS-based implementation of Kimi Delta Attention kernels. It delivers a 1.72×–2.22× prefill speedup over the flash-linear-attention baseline on H20 and works as a drop-in backend for flash-linear-attention.
Source: Kimi · x.comPublished · added here