Billy Cao
> > > @neoOpus all of his errors come from the requests library. It is fair to say that hardware does not matter here. He will still get the same...
Same issue with torch 11.3+cu117 on Windows.
Problem is caused by extra line prepended to the return data: data:image/png;base64,
I opened PR for a fix https://github.com/AUTOMATIC1111/stable-diffusion-webui/pull/3974
Update: this is fixed separately by @AUTOMATIC1111 himself in https://github.com/AUTOMATIC1111/stable-diffusion-webui/commit/198a1ffcfc963a3d74674fad560e87dbebf7949f
I can workaround by adding `t = t.to('cpu')` above that error line, but it takes up more RAM and this should still be a bug
Oh was this fixed in a recent PR? Because my version was from 3 days ago
> > Same issue, on a fresh environment with a cloud-based machine. Edit, resolved by clearing up torch package version mismatches/reinstalling torch. > > @rapidcopy Can you elaborate more on...
> I don't understand the code, but my GPU is currently running at 100% as I run textual inversion training, so I'm not sure. Oh, my mistake, I mixed up...
> According to `nvcc --version` I am on CUDA 11.7 too (updated after attempting to use xformers), so I have no idea why anything is working on my end Your...