Initial version. Trained on 10 images, 600 steps, 6 epochs.
LoRA for generating consistent images of Maya, an Asian woman with vitiligo skin patterns. Trained on SD 1.5.
