LoRA created from 100 images with art by 乱視と君と。(santa):Recommended weight 0.7 to 1.0 depending on your model and prompt.No trigger word is needed, but you can use 'ranshi_kimi' if you want.A nice style but not strong enough to win over some base models. Trained mostly on monochromatic manga images, part of the visual style and some characters are clearly easier to represent.

Files with SHA256: 1f7c58f5700e6cd72c0f8185d81493e1c1de90a9aefe99eec2eab0eaf1931858