A few nice images and video got hidden automatically — just turn on all the Browsing Levels to see them.
V2
Recommended weight: 0.9-1（0.9 couldn’t be better）
Recommended settings
Sampler: Euler
Schedule type: Simple
Steps: 30
CFG: 5
Hires. Fix: R-ESRGAN 4x + Anime6B
Magnification: 2
Prompts
Prompt Prefix:
masterpiece, best quality, newest, absurdres, highres, safe, s2d3o, 1girl, solo, looking at viewer, bare shoulders, hair between eyes, smile, blushNegative Prompt:
nsfw, worst quality, old, early, low quality, lowres, signature, username, logo, bad hands, mutated hands, mammal, anthro, furry, ambiguous form, feral, semi-anthroV1
Recommended weight: 0.8-0.9（From testing, 0.8 seems to be the optimal choice — it’s more stable）
Recommended settings
Sampler: Euler A
Schedule type: Simple
Steps: 30
CFG: 6~7（It feels like the colors in 6 are richer）
Hires. Fix: R-ESRGAN 4x + Anime6B
Magnification: 2
Prompts
Prompt Prefix:
s2d3o, 1girl, solo, looking at viewer, bare shoulders, hair between eyes, smile,Negative Prompt:
nsfw,bad quality,worst quality,worst detail,sketch,censor,Description
This version train on NoobAI-XL (NAI-XL) V-Pred-1.0. I also optimized the dataset and improved the tagging strategy so the tags align better with the image features for the AI to learn,at the same time, i’m also exploring new training strategies