This experimental LoRa is meant to boost a base model's illustrated art style knowledge, color, poses, and detail accuracy. Works best on anime or furry models.

Use a strength between 0.5 to 0.7