nodes/hao-ai-lab-FastVideo
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

[perf] Add FA4 CuTe backward support for VSA-256 (#1639)

Browse Source 
Co-authored-by: Hyunsung Lee <hyunsungl@sizigistudios.com>
Co-authored-by: alexzms <3036648523@qq.com>
This commit is contained in:
Hyunsung Lee
2026-08-19 11:46:49 -07:00
committed by GitHub
co-authored by Hyunsung Lee alexzms
parent 8537dcd6de
commit 00338aa9ca
13 changed files with 929 additions and 140 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docker/Dockerfile
+1 -1
View File
@@ -68,7 +68,7 @@ ARG FLASH_ATTN_WHEEL_RELEASE_ARM64=https://github.com/mjun0812/flash-attention-p
# cutlass-4.4 `cute.core.ThrMma` API, which crashes on the cutlass-dsl 4.5 that
# flashinfer/quack pull in. After the wheel install we overlay this cutlass-4.5-safe
# upstream cute (flash-attn-4) so the image runs FA4 instead of the FA2 fallback.
ARG FA4_CUTE_REF=82d6441eec5d4dfec120153db2c0145ae855a083
ARG FA4_CUTE_REF=14c377950125c70b7a9dabf9c561fca53715ac7d
# Provided automatically by BuildKit/buildx (e.g. "amd64" / "arm64") and used to
# select the prebuilt flash-attn wheel. Empty under a plain `docker build` without