This can greatly reduce the inference time of recent SOTA video models like Wan2.1. Any plans to support video models?

Open MeiYi-dev opened this issue 8 months ago • 0 comments

Apr 25 '25 14:04 MeiYi-dev