Text-to-Image
Diffusers
Safetensors
StableDiffusionPipeline
stable-diffusion
Inference Endpoints

error!!!

#18
by mary34 - opened

download 768-v-ema.ckpt error!!!!!

I also have an issue im getting a file smaller 4.9 GB when i tried to download it... 300MB seems to me miss

I get an ERROR just selecting the Model in webui. Then I get this: runtimeerror: expected all tensors to be on the same device, but found at least two devices, cpu and cuda:0! (when checking argument for argument index in method wrapper__index_select) I can run all the other models, just not SD 2.0 for some reason...

I change for another yaml and it works

Of course, I use webui stable diffusion, the Model in webui... what a pity I want to see what SD 2.0 is like? some realistic photos.
where did you change the yaml file from?

deleted
This comment has been hidden

I get an ERROR just selecting the Model in webui. Then I get this: runtimeerror: expected all tensors to be on the same device, but found at least two devices, cpu and cuda:0! (when checking argument for argument index in method wrapper__index_select) I can run all the other models, just not SD 2.0 for some reason...

Apparently using --medvram or lowvram causes this ?

Has anyone had this issue and got it fixed? I have no idea why it only works for some models but failed on the same error for realistic ones or some other models of its kind.
Please can anyone help on this? Thanks so much!

Sign up or log in to comment