I Made Stable Diffusion XL Smarter by Finetuning It on Bad AI-Generated Images

pbjtime wrote: post #4

brianjking wrote: post #3

Earlier quoted context omitted.

What? XL is the current version of Stable Diffusion.

It's already on version 40?

Extra large

Re: I Made Stable Diffusion XL Smarter by Finetuning It on Bad AI-Generated Images

#6

Post by sschueller » Mon, Aug 21, 2023, 4:37 PM UTC

kwhitefoot wrote: post #2

> XL Extra Large? 40 times?

1024 x 1024 instead of 512 x 512.

Re: I Made Stable Diffusion XL Smarter by Finetuning It on Bad AI-Generated Images

#7

Post by jfoutz » Mon, Aug 21, 2023, 4:41 PM UTC

bckr wrote: post #5

pbjtime wrote: post #4

Earlier quoted context omitted.

It's already on version 40?

Extra large

It’s a Roman numeral joke.

Re: I Made Stable Diffusion XL Smarter by Finetuning It on Bad AI-Generated Images

#8

Post by msp26 » Mon, Aug 21, 2023, 4:52 PM UTC

>A minor weakness with LoRAs is that you can only have one active at a time

Uh this isn't true at all, at least with auto1111.

Re: I Made Stable Diffusion XL Smarter by Finetuning It on Bad AI-Generated Images

#9

Post by rabuse » Mon, Aug 21, 2023, 4:52 PM UTC

Creating art with stable diffusion has become such a fun hobby of mine. The difference between SD 1.5/2.0 and SDXL is massive, and it's impressive how quickly the quality is improving with this stuff.

Re: I Made Stable Diffusion XL Smarter by Finetuning It on Bad AI-Generated Images

#10

Post by politelemon » Mon, Aug 21, 2023, 4:53 PM UTC

Please consider posting the LoRa on civitai.com as well as the stable diffusion Reddit.

These results look pretty good, looking forward to trying it out. I hadn't realized that the generative images buzz was dying out, since I'm using it regularly I guess it is always in buzz for me.