infer issue of GoatWu/Self-Forcing-Plus

[solitaryTian]

Thank you very much for your work. The acceleration lora you trained has worked extremely well.

I learned from huggingface that you trained using the GoatWu/self-Forcing-Plus repository. But when I trained the text-to-video acceleration model (a single.pt file) using the self-Forcing-Plus repository, I encountered some problems during inference.

I wonder if I should use lightx2v or self-Forcing-Plus for infering video? When using lightx2v, I found that the code only supports.safeTensors files. The self-Forcing-Plus repository does not seem to support inference for the 14B model either