This is an experimental LoRA created while preparing for my next checkpoint training. Designed primarily for inpainting, it allows for creative edits and refinements. While you can try using it for non-inpainting tasks, keep in mind that it may distort facial features - especially when combined with character-focused LoRAs.

This is just a test version, so don't expect perfection yet! The next iteration will include more diverse piercings (not just facial ones) and improved flexibility. Feedback is always welcome as I continue experimenting and refining this concept.

Enjoy testing it out and let me know how it works for you
It works good with default Flux.dev and pairs beautifully with my UltraReal checkpoint as well, but slightly better then flux.dev, so feel free to experiment

Settings: Dpmpp2m + beta scheduler, Denoise 0.8-0.9
P.S: also attached workflow in models manager