Trained from mostly anime screencaps. Due to limited dataset, if generated image appears stiff, consider adjusting the weight downwards and utilizing img2img or ADetailer for enhanced results. Recommended weight range: 0.8 to 1.
Update: v2.0 added ahoge dataset.
Prompt below is for V2.0 for V1.0 please click desired version for details.
Base:
arcana, white hair, gold eyes, blunt bangs, hair tubes, short hair, long locks
arcana, white hair, gold eyes, blunt bangs, hair tubes, short hair, long locks, ahoge
Outfit:
goddess outfit, white headwear, cape, rope, white sleeves, long sleeves, cross, gold trim, forehead jewel