Hello again my high culture fellows. I came back after chill month with new LoRA.

I noticed that Illustrious and NAI not good in such sexy pose so I decided to make it.

Recommended model for this LoRA:

For V-pred version: - NoobAI-XL V-pred 1.0 or anime mixes.

For Epsilon-pred version - NoobAI-XL Epsilon 1.1 , Cyberfix versions or NAI Epsilon mixes.

For Illustrious 0.1 version use Illustrious 0.1 or Illustrious mixes.

Please, read to prevent crap AI slop making:

Main tags:

Legs on table, on chair, chair, table, sitting

What in frame:

full body, feet out of frame

Legs position:

crossed ankles, crossed legs, legs together, spread legs

Important angles:

straight-on, three quarter view, from side

Tags for feet:

1) Any footwear, like "high heels, loafers" or "no shoes" if char wear thighhighs, socks or pantyhose.

2) Or naked legs: "barefoot, soles, toes, foot focus"

Extra:

You can add "3d" or "photorealistic" for 3d render or photo.

NoobAI-XL quality settings:

Recommended generation settings:

Euler A or Euler SGM Uniform, 30 steps.

CFG 4-5! (NOT MORE FOR BOTH VERSIONS).

For V-pred only: rescale CFG 0.3.

Highres fix:

0.25-0.5 denoising strength, 12-20 hires steps,

Upscalers: Foolhardy Remacri 4x (good and fast, very popular) or DAT 2 x2 (VERY powerful tile upscaler but slow. His main advantage - saving many small details).

Also you can try HAT-S SR 2x - it is not so fact as esrgan (foolhardy remacri) but not so slow as DAT and it's also good modern tile upscaler as DAT).

Adetailer:

Denoising strength 0.25 - 0.3. Detection model confident threshhold: 0.7-0.8.

Inpaint only masked padding pixels - 120-140 pix for face models, 156-256 for hands models.

Please don't forget to rate my LoRA. This motivates me to do more!

https://tensor.art/u/604375505701151021