@wangshankun @gushiqiao @helloyongyang @fuheaven @Lubenwei-nb123 @helloyongyang @GoatWu Hi, is there a way to support the newly released ByteDance Helios models for acceleration purposes? (An 14B autoregressive diffusion video generation model that faster than 1.3B ones) Code: https://github.com/PKU-YuanGroup/Helios Page: https://pku-yuangroup.github.io/Helios-Page/ https://github.com/user-attachments/assets/65e4d770-fd1e-4834-9732-6156846f69d2
@wangshankun @gushiqiao @helloyongyang @fuheaven @Lubenwei-nb123 @helloyongyang @GoatWu Hi, is there a way to support the newly released ByteDance Helios models for acceleration purposes? (An 14B autoregressive diffusion video generation model that faster than 1.3B ones)
Code: https://github.com/PKU-YuanGroup/Helios
Page: https://pku-yuangroup.github.io/Helios-Page/
T0mcBl4hGMFHPlVHjqOxE.mp4