ice051128
ice051128
Anaconda is literally bloatware. Just follow the instructions and get python 3.10.6 instead and let webui install the packages on first launch
The easiest fix is to use an older version of launch.py, one of his commits broke it completely
> [SOLVED] this is happening when i press "send to img2img" or "send to inpaint" from the Image Browser tab. ..but same buttons pressed in txt2img tab works fine. :(...
> > Did not fix here. > > > same. ERROR is still there > > Pull Request #51 fixed the issue for me, but I had actually solved it...
> > A simple parameter change from #39 works flawlessly. What's the point of changing the whole function? > > doens't work here, i just checked the line in #39...
> Adding torch.backends.cuda.matmul.allow_tf32 = True makes creation faster. When tested, the speed is slightly faster than xformers, and you can enjoy the miracle that the speed is about 2 to...
> You can think of the principle as replacing fp32 with tf32 to increase the calculation speed. Any impact on output? Can u reproduce old images with less numerical accuracy?...
> This only works on RTX 3000 and newer cards, but as AUTO said, it is already true now, so this PR might not be needed Why does this PR...
> > If no one tests it I'll do it when I get home. > > I've tested on my 3060 again, SDv2.0, 768x768 with `--xformers` without `--no-half` > >...
> > Have u tested the fix without --no-half-vae? > > I didn't, because without `--no-half-vae` it will _lower_ the spike at the end of generation. But in my case,...