LoRA trained on the artwork of TheOfficialPit. V2 is here!

It took some time to improve the original dataset and produce the new synthetic images, but you can now enjoy a better version of ThePit Style Anima LoRA.

V1 was trained on a mix of non-synthetic images from The Official Pit and images generated with Illustrious to fill gaps in the dataset’s coverage.
In V2, all those synthetic images have been replaced with either newer work from the artist or high-quality synthetic images generated with Anima.

I use it at a strength of 1.20–1.50.

V3 is planned because I feel that V2, while close, still needs a little more work to perfectly match ThePit’s style while remaining flexible with prompts.

If you get good results, feel free to post them on the model page.
Feedback is welcome, especially if you notice prompts where the style does not apply well.