My images today

https://i.rdrama.net/images/1711577445652507.webp https://i.rdrama.net/images/17115774474133172.webp https://i.rdrama.net/images/17115774488236823.webp https://i.rdrama.net/images/17115774497186363.webp https://i.rdrama.net/images/17115774513396132.webp https://i.rdrama.net/images/17115774523075747.webp

playing around with trying to get SDXL to actually do different styles, the pony model is pretty resistent to style changes

15
Jump in the discussion.

No email address required.

the pony model is pretty resistent to style changes

Autismmix seems to do better. I've had good luck with manga, b/w monochrome, and a few of the artists I like.

Jump in the discussion.

No email address required.

I call autismmix pony but yeah I use that.

I just find it far worse compared to 1.5 at changing styles. https://i.rdrama.net/images/17115791827863367.webp

Same data set making the loras but pony model takes the change much "weaker" however I tweak it

Jump in the discussion.

No email address required.

The Asanagi, Oda-non, and a few others I've gotten off Civitai do fine.

Jump in the discussion.

No email address required.

Link copied to clipboard
Action successful!
Error, please refresh the page and try again.