3D ART STYLE NoobAI-XL Epsilon V2

LORA
Original


Updated:

85

New version of 3D ARTSTYLE LORA is on NoobAI-XL Epsilon now!

BETTER 3D STYLE, BETTER DATASET, BETTER TRAINING SETTINGS! NO RANDOM NOISE ANYMORE!

LoRA specializes in creating artistic 3D images by adding lighting effects and knows a lot of 3d game characters.

!!! FOR NOOBAI-XL V-PRED (CYBERFIX) USE THIS LORA !!!

USE THIS FOR NOOBAI-XL EPSILON (1.0 OR 1.1). BEST MODELS TO USE 3D LORA:

NoobAICyberfix1.1 perpendicular

NoobAiCyberfix v 3.0 (v.1.1)

NoobAI-XL Epsilon (v. 0.75, v 1.0 or 1.1)

I know that according to strange civitai policy cyberfix is not available in site generator. I hope it will be in future and also I'll try available checkpoint-mixes which also can be used for 3D.

Please don't use it on SDXL 1.0 base, PonyDiffusion or pure Illustrious 0.1 because your result can be worse!

NoobAI-XL tags:

Main tag:

3d

Body effects and tags:

shiny skin, wet, oiled, toned, curvy, plump, muscular, tan, tanlines, cellulite

sidelighting, backlighting, underlighting, bloom, shadow, light paricles, sunlight, dappled sunlight, light rays, moonlight

To prompt known franchise characters use prompt order: {character name}, {francise name}, other tags. (FOR VERSION 2 DON'T NEED FRANCHISE TAG)

All characters in my LoRA are tagged with danbooru tags! So use tags from this site: https://danbooru.donmai.us/

My preview can help.

NoobAI-XL quality settings:

Put "masterpiece, best quality, good quality" in positive and put "low quality, worst quality" in negative.

Use euler, cfg 4-6. Don't use higher CFG (without dynamic thresholding or something like that).

If you have over-constrasted picture and cfg 4-6 didn't help you can try to put in negative "colorful".

More in my article.

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

Upscaler / adetailer settings:

Hires.fix: Upscaler foolhardy_remacri (download and put in "ESRGAN" folder in "models" if you are using automatic1111 or forge).

Hires steps: 16-18

Denoising strength: 0.3

Adetailer: denoising strength 0.3. Detection model confident threshhold: 0.7-0.8.

Best sampling method is euler / euler a, 25-30 sampling steps, You don't need any DPM++

https://tensor.art/u/604375505701151021

Version Detail

Illustrious
For download visit my civitai page.

Project Permissions

    Use Permissions

  • Use in TENSOR Online

  • As a online training base model on TENSOR

  • Use without crediting me

  • Share merges of this model

  • Use different permissions on merges

    Commercial Use

  • Sell generated contents

  • Use on generation services

  • Sell this model or merges

Related Posts