#

fused-mla

Here is 1 public repository matching this topic...

ffpa-attn

xlite-dev / ffpa-attn

📚FFPA(Split-D): Extend FlashAttention with Split-D for large headdim, O(1) GPU SRAM complexity, 1.8x~3x↑🎉 faster than SDPA EA.

cuda attention sdpa mla mlsys tensor-cores flash-attention deepseek deepseek-v3 deepseek-r1 fused-mla flash-mla

Updated May 10, 2025
Cuda

Improve this page

Add a description, image, and links to the fused-mla topic page so that developers can more easily learn about it.

Curate this topic

Add this topic to your repo

To associate your repository with the fused-mla topic, visit your repo's landing page and select "manage topics."