First test of the style!
A LoRA style trained on a mixed dataset of 2D anime girls, focusing on soft lighting and vibrant colors.