hao-ai-lab/flash-attention-fp4

SkillsMP has collected 1 skills from hao-ai-lab/flash-attention-fp4. Open a skill to review its source and details.

hao-ai-lab profile View on GitHub

Latest recorded source activity: July 23, 2026 at 04:17

SkillsMP catalog refreshed: September 10, 2026 at 08:02

skills collected: 1

GitHub stars: 53
GitHub forks: 7

Skills in this repository

Software Developers

1 occupation categories · 100% classified

skill

occupation

description

updated

kernel-bottleneck-tuning

Software Developers

Workflow for finding and fixing instruction-throughput bottlenecks in warp-specialized CuTeDSL kernels (FA4-style). Use when a kernel underperforms its roofline and you need to decide between reordering, eliminating, or re-pipelining work.

2026-07-23

Showing 1 of 1 collected skills.