Introducing OctoMix.
This Frankenstein of a merge is between Illustrious 2.0, WAI-Illustrious-SDXL, a checkpoint I trained locally that kind of got to where I wanted but wasn't quite there, and a LoRA I trained locally to further push the aesthetic I wanted.
I've always wanted to do some things with AI and finally got a good enough PC to do it locally. It's probably not that special, but I hope you enjoy it!
I'm still not too familiar with all the technical stuff but I'll try to help if I can.
The VAE is already integrated.
***Prompting Suggestions***
Positive Prompt: [population/type of scene], [character/character LoRA trigger], [description from head to toe], [location], [camera angle], masterpiece, best quality, absurdres, newest
Example: 1girl, Tifa, sundress, outdoors, upper body, masterpiece, best quality, absurdres, newest
I think it works better when the positive is also under 75 tokens.
Negative Prompt: lowres, worst quality, bad quality, bad anatomy, sketch, jpeg artifacts, signature, watermark, artist name, old, oldest, multiple views
Steps: 20-30
CFG scale: 7
Sampler: Euler a
Scheduler: Karras
***Disclaimer***
This model uses the Fair AI Public License 1.0-SD license.