The purpose of this model is to add an increased range of diversity to your basic generations.

When prompting for a person without a specific ethnicity specified, this model will make the outputs more diverse. It works very well at a distance, but has some issues with close-ups of people. To be improved in the future.

Note:

The diversity is not limited to people. The model is also trained on more diverse locations. A livingroom is not the same everywhere in the world.

Hopefully this model unlocks and makes you things you haven't seen or experienced before.

Generation Settings

It is flux, anything seems to work fine for me. I have no particular recommendations to give for now.