Saturno | IL

Hi there... bringing another attempt at a good merge. Hopefully, this one seems to be giving out amazing results with a few bad ones in the mix, but overall good, especially for more artistic styles and artist tags. I've tried a lot of prompts from different creators to see how it fares against this merge, and some of them are amazing, while some of them failed miserably. But overall, here we are.

Here is a list of all the models merged into this:

• ​🇳​​🇴​​🇴​​🇧​​🇦​​🇮​-​🇽​​🇱​ - by L_A_X

• ​🇮​​🇱​​🇱​​🇺​​🇸​​🇹​​🇷​​🇮​​🇴​​🇺​​🇸​-​🇽​​🇱​ ​🇻​2.0 - by ONOMAAI

• ​🇵​​🇱​​🇦​​🇳​​🇹​ ​🇲​​🇮​​🇱​​🇰​ 🌿 - by Ocean3

• ​🇳​​🇦​​🇮​​🇽​​🇱​-​🇲​​🇲​​🇲​​🇮​​🇽 - by xiaofan941297

• ​🇴​​🇧​​🇸​​🇪​​🇸​​🇸​​🇮​​🇴​​🇳​ 🇮​​🇱​​🇱​​🇺​-​🇽​​🇱​ - by rqdwdw

Recommended Settings

Recommended settings (so far): Euler, DPM++ 2M Karras, Restart. I only tested 28~33 steps accordingly (but might work at lower steps, haven't tested it out). CFG around 3 to 4.3 was the max I used (I usually keep the CFG at either 3.6 or 4.1).

Tags

Some positive tags I've used: masterpiece, best quality, amazing quality, very aesthetic, absurdres, highres.

Negatives (taken from Plant milk's gens): bad quality, worst quality, lowres, jpeg artifacts, bad anatomy, bad hands, multiple views, signature, watermark, censored, sketch, flat color, ugly, fat, muscular, abs, blurry eyes, wrinkled skin.

VAE Issues

In case the results are coming out muted and filled with white dots, that's a VAE issue. I'd recommend using illustrious-xl-v20, or any illu VAE you prefer, but this is the one I tried and it worked.

Support

If you want to support me, you can either donate, drop a like, or follow me for future updates; either one helps. Thank you so much, every single buzz or support that I receive is going straight into my LoRAs.

Licenses

SDXL - CreativeML Open RAIL++-M

Illustrious - https://freedevproject.org/faipl-1.0-sd/

NoobAI-XL - /model/license/1140829

Also, Fair AI Public License 1.0-SD