Download
1 variant available
9120 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 9
1410 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 9
2340 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 9
(72)
Jan 2, 2026

4910 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 9
7.7K0 1 2 3 4 5 6 7 8 9.0 1 2 3 4 5 6 7 8 9K
76.4K0 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 9.0 1 2 3 4 5 6 7 8 9K
License:
Apache 2.0Ok so this is my first attempt in Training on ZIT - originally I wanted to wait for the full model, but since it will take longer than expected, I wanted to give it a try, I am really impressed by the capabilities of ZIT, even though I just used a small Dataset, without any optimization etc.
I again tried to capture the "retro..." feel of the late 70s and 80s magazines. I think this one is the best from all my attempts. ZIT is really on another level.
The Lora adds also a more realistic look and more character diversity, people look more convincing. Works also well with a Img2Img Workflow.
Important Notes: Use this text before your main prompt, to enhance the effect
Retro_zit.
adult content from the 80s, muted colors with low contrast. subtle sepia tint, high film grain. Scandinavian adult magazine.
natural skin, with subtle natural imperfections to keep a realistic depiction of a human body.
Then add your prompt.
I used Euler and Simple
-> keep the strength between 0.4-0.7, I mostly used .6




