Software Development

Flashkda Delta Attention

High-performance Kimi Delta Attention CUDA kernels built on CUTLASS for efficient recurrent state inference

Publisher / source
reason-machines/trending-skills (found on skills.sh)
Upstream record
reason-machines/trending-skills
Version
Version 1
Published
Publication review
Approved for publication
Content fingerprint
988d12b470863c531b2b717edbb60fe76adb4a2faefa5d3d61bb99d1d7f54ba2