cian0

I can see that this happens on Android browsers, I am using Firefox/Brave browser/Chrome on my tablet and I can confirm that it doesn't work (both click to add components...

There's this conversion script but haven't really tried it yet https://gist.github.com/laksjdjf/dc43fc6c5cb4d9000fbbf18515f1e4d8

Oh nice, but how about the textual inversion that gets trained? And is it compatible with dev branch?

Same bug, I tried the fix here by changing the pytorch source directly and set allow fp16 = True , the training went through but the model only outputted black...

To add, I also tried reverting the diffusers version, I am using Shivam's repo which is a fork of diffusers, he merged the latest version of diffusers which caused issues....

Thanks! I figured this is 2.1 sd, any idea how i can merge existing models into this one?

so if I use images of trains, non anime persons, animals, it is still possible?

But it is configurable under pilot/const/code_execution.py right?

Would it be possible to use other base models as what's being done in SD WebUI a1111?