DoRA is trained on portraits and images of clothing from the game Disco Elysium.

DoRA weight: I recommend using the DoRA weight between 0.7 and 0.9. I personally use 0.85.

Recommended model: DoRA is trained on the base SDXL 1.0 model, so I think it's okay to use this DoRA on any model that is based on SDXL 1.0 (except anime models).

Clip skip: The clip skip of the DoRA is 1.

DoRA triggers on the words: illustration portrait, abstract illustration portrait, multicolored (or any other color) background and illustration of a dress (or any other clothes, accessories, etc).

Use portrait aspect ratio when generating. I use a resolution of 768x1024 for portraits and 1024x1024 for clothing.