This website requires JavaScript.
Explore
Help
Sign In
anodyine
/
vllm-omni
Watch
1
Star
0
Fork
0
Code
Issues
4
Pull Requests
2
Actions
Packages
Projects
Releases
Wiki
Activity
Files
2da9d28f1b378eb63fcc1868dd116e3a044af8cb
vllm-omni
/
tests
/
diffusion
/
attention
T
History
tzhouam
2da9d28f1b
Enhance reproducibility in Ulysses attention tests by replacing current_omni_platform.seed_everything with torch's manual_seed and adding support for CUDA devices. Introduce get_supported_tasks method in AsyncOmni class to dynamically determine available tasks based on stage capabilities.
...
Signed-off-by: tzhouam <
tzhouam@connect.ust.hk
>
2026-03-03 09:45:51 +00:00
..
test_attention_sp.py
Enhance reproducibility in Ulysses attention tests by replacing current_omni_platform.seed_everything with torch's manual_seed and adding support for CUDA devices. Introduce get_supported_tasks method in AsyncOmni class to dynamically determine available tasks based on stage capabilities.
2026-03-03 09:45:51 +00:00
test_flash_attn.py
[XPU] Enable FLASH_ATTN on XPU (
#1332
)
2026-02-26 20:28:09 +08:00