Makes the subject more Asian and the whole image less saturated.
A proof of concept for Flux Lora training.
Training Set: 11 stable diffusion generated images.
Training Steps: 1100.
Recommended strength: 0.7 - 1.0 .
Increase the strength if the subject is not Asian enough.
Type A has better anatomy than Type B, and is a little bit more nsfw friendly. ;)