B200 Flash Attention4 Planner

Use when the user wants to design or extend a FlashAttention-style forward kernel on B200/Blackwell, involving the two MMAs QKᵀ and PV, online softmax, S/P/O in TMEM, warp roles, causal mask, GQA, tile scheduling, or final normalization. Outputs the algorithm state, tile graph, barrier graph, and validation plan. Not for cases that only use off-the-shelf framework operators, where the full backward is not yet defined, or for ordinary dense GEMM.

mirage-project Updated 2.6k repo stars

File contents

mirage-project/mirage commit 426bb0eeb5

Frequently asked questions

npx get-skill add mirage-project/b200-flash-attention4-planner