HyVideoTextEncode - All input tensors need to be on the same GPU, but found some tensors to not be on a GPU: [(torch.Size([1, 8388608]), device(type='cpu')), (torch.Size([262144]), device(type='cpu')), (torch.Size([4096, 4096]), device(type='cpu'))]

[chaoscode]

I don't know why this is an issue but it's detecting the cpu instead of the GPU on docker. The GPU is attached to the docker fine as I have mochi working on the same docker. Can anyone provide any guidance on how to fix this error?

[FlareP1]

Possibly had a connected error see Issue 80, In my case the problem was that the --lowvram flag on launch was causing issues with memory allocation and forcing some tensors onto the CPU.

Are you using the --lowvram on the launch of comfyUI? If so you could try to remove it?

[chaoscode]

I will look into it. I don't think it's a switch because I'm using the official comfyui docker from github on latest. Nothing special.

[chaoscode]

Yeah, I tried a non-lowvram workflow and it's just crashing the container. I think I'm just going to move on from this model. Thanks for the help.