Make sure to use UI that supports v-prediction, otherwise you will get a noisy blob!
See about version for credits!
Main parameters:
Steps: 28
CFG: 4
Sampler: euler a (euler for v6 and lower)
Scheduler: normal
Im also using CFGZeroStar and sometimes Self-Attention Guidance: default on difficult backgrounds)
Positive prompt:
number of characters (1girl/1boy), character name, artist prompt, scene/environment/camera angle, action, expression, items, quality prompts (masterpiece, best quality)some heavily stylized characters might require prompt: realistic
you can also try to put nose into faceDetailer
Negative prompt:
lowres, worst qualityurl against watermark
lineart, sketch, dot nose for realism
Enjoy..
Description
Revisited thanks to @Uberlord, tamed the models for more prompt adherence, darkness and overall stability.
Models used, go like them!
FAQ
Comments (3)
What makes v10 and v11 different from each other? It's a fantastic model, but what makes v11 stand out from v10? Better prompt adherence? Better quality? More data?
I have typed it into the about version.. xddd
But again then, v11 is made using lower step counts for LoRAs I used and I have tested it much more since I had more time on my hands. Played with custom block weights before merging for hours to find the best general use values.
Essentially: style1800steps>style1300steps, slimetranslucency700steps>slimetranslucency300steps, 2DN model from 33% to 20%.
@blinkdotleh I have noticed that v11 is just a smidge softer with a touch of blur. I'm not sure if that was the intended goal, but it's still a fantastic model