Home Artists Posts Import Register

Downloads

Content

🍑I wonder what it's possible to read in that Braille: Yummy!

🖼️ See some examples 🖼️

Unlike the last model, OnionBooty (trained with black butts), this time I trained with PAWG models. This way stretch marks and moles look much more realistic.

⚖️LoRA prompt weight: 0.6~0.8
⚙️~99% effectiveness
🏫Model used for training: CyberRealisticV3.2

🔍Recomendations:

  • CFG ~9
  • Steps >=35
  • All Samplers

I recommend start your prompts with:

[photo type] [nationality], beautiful face, [nude or clothing], [pose and position], [woman characteristics], [ass characteristics] …


  • This model is very flexible, you can change prompt order if you want.
  • To nudify: nude, naked
  • Leggings work with “[color] leggings” prompt
  • Recommended body characteristics: huge ass, thick thighs, wide hips, small waist

Recommended starting neg with:

(plump, fat:0.9)...

🌟Tutorial for improving results.

LoRA details and tendencies (+resolution):

  • I achieved good results with these neg embeddings: CyberRealistic_Negative-neg and FastNegative. But you can use normal neg prompts.
  • This Lora can cause some little face problems, I recommend inpaint them without the LoRA prompt or use Adetailer, check the improvement link to learn how you can use that.
  • The LoRA sometimes fattens people: use "plump", "fat" on neg and "skinny" on pos.

Inside the folder I left some example images and tags used for training, you can use this on your prompts.

Comments

No comments found for this post.