biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

Use official xformers-0.0.33 built for PT 2.9 (#28600)

Browse Source 
Signed-off-by: Huy Do <huydhn@gmail.com>
Co-authored-by: Roger Wang <hey@rogerw.io>
(cherry picked from commit c33b87e777)
This commit is contained in:
Huy Do
2025-11-12 22:48:53 -08:00
committed by Kevin H. Luu
parent caf412e593
commit 1326f17492
1 changed files with 1 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
requirements/cuda.txt
View File

@@ -9,7 +9,6 @@ torch==2.9.0
torchaudio==2.9.0
# These must be updated alongside torch
torchvision==0.24.0 # Required for phi3v processor. See https://github.com/pytorch/vision?tab=readme-ov-file#installation for corresponding version
# Build from https://github.com/facebookresearch/xformers/releases/tag/v0.0.32.post1
xformers==0.0.33+5d4b92a5.d20251029; platform_system == 'Linux' and platform_machine == 'x86_64' # Requires PyTorch >= 2.9
xformers==0.0.33; platform_system == 'Linux' and platform_machine == 'x86_64' # Requires PyTorch >= 2.9
# FlashInfer should be updated together with the Dockerfile
flashinfer-python==0.5.2