Yuxiong WU

icon location Hangzhou, China icon location Let's build the future with AI

Results 2 issues of Yuxiong WU

wan_video_dit issue with flash_attn 3

3 comment

Traceback (most recent call last): File "/data/yuxiong/DiffSynth-Studio/examples/wanvideo/wan_fun_InP.py", line 65, in video = pipe( ^^^^^ File "/data/yuxiong/miniconda3/envs/torch26/lib/python3.11/site-packages/torch/utils/_contextlib.py", line 116, in decorate_context return func(*args, **kwargs) ^^^^^^^^^^^^^^^^^^^^^ File "/data/yuxiong/DiffSynth-Studio/diffsynth/pipelines/wan_video.py", line 349, in __call__...

Does Wan VACE support fine-tuning, or LoRA integration? Or we can just train LoRA for Wan 2.1 and apply to VACE

1 comment