ValueError: Cannot load E:\tools\ComfyUI_windows_portable\ComfyUI\models\diffusers\stable-video-diffusion-img2vid-xt-1-1 because decoder.conv_in.bias expected shape tensor(..., device='meta', size=(64,)), but got torch.Size([512]).

[gladmustang]

what's the reason of this ?

[kijai]

Can you try updating the nodes now and running again?

[gladmustang]

it works now, thanks