Make sure to use UI that supports v-prediction, otherwise you will get a noisy blob!
NoobAI-XL (NAI-XL) based.
Main parameters:
Steps: 28
CFG: 3-5
Sampler-scheduler: euler/euler a-normal, DDPM-exponential (better noses)
CFGZeroStar and sometimes Self-Attention Guidance: default for difficult backgrounds
Try to BREAK prompts into parts by 75 tokens max, this goes for most models..
Positive prompt:
number of characters (1girl/1boy), character name, artist prompt, scene/environment/camera angle, action, expression, items, quality prompts (masterpiece, best quality, absurdres)some heavily stylized characters might require prompt: nose, realistic.. you can put nose into Adetailer
Negative prompt:
worst quality, low qualityeven longer negative should work, but I prefer simplicity.. tags worth noting include sketch, abstract, url
Enjoy..
Description
Focus on prompt adherence and anatomy rather than style.
FAQ
Comments (3)
Hi Boss, please, can you explain in two word what exactly difference between illustrious and noobai?O_O
I tried googling it, but it was all math and formulas((
Noobai is further trained illustrious v0.1 has more danbooru knowledge, better colors imo, but untamed styles, which I tried to fix.
@blinkdotleh Yeah, the model turned out great! I also noticed that the noob has less fantasize then illustious models. Although Im not sure









