What is this?
This experimental LoRa is meant to boost a base model's illustrated art style knowledge, background accuracy, color, poses, and detail accuracy. Works best on anime or furry models.
Note: May not work as intended on Civitai's site generator or certain UIs, as this has been an issue for some users with the LoRas I build using ComfyUI for unknown reasons. It works for sure in ComfyUI as I have compared with/without the LoRa many times with different settings.
Usage Tips:
Use a strength between 0.5 to 0.7