MoonshotAI open-sources FlashKDA, a high-performance Kimi Delta Attention kernel
MoonshotAI · github · 2026-07-29
MoonshotAI releases FlashKDA, a high-performance Kimi Delta Attention kernel
MoonshotAI has open-sourced FlashKDA, a set of high-performance Kimi Delta Attention kernels.
- Language: CUDA
- Repo size: 889 stars and 216 stars gained today
- The project appears aimed at low-level performance optimization for Kimi-style attention
More from Infra
- Tencent open-sources AngelSpec, a speculative decoding framework with up to 2.4x speedup — teortaxesTex · 2026-07-29
- Kimi K3 pitch says firms spending over $20k a month on APIs may be better off self-hosting — ZeYanjie · 2026-07-29
- Bloom and Seagate ride AI demand while Vertiv’s broader portfolio underperforms — TiernanRayTech · 2026-07-29
- llama.cpp merges speculative decoding support for GLM-5.2 — YPSONDESIGN · 2026-07-29
- ChipAgents raises $60M Series A2 and says ARR is up 6x in H1 2026 — WilliamWangNLP · 2026-07-29
- NeurIPS 2026 workshop seeks papers on decentralized foundation model training — peter_richtarik · 2026-07-29