Trigger word: ImHigh

Dataset: Random Cool Images

Training: New settings I've never tried before

It was a great experience seeing how a bunch of random cool images could make a cool multi-purpose LoRA, and the results turned out much better than I expected.

Training image resolutions: 673x1200 and 736x1308.

Prompts worth trying:

Expressive characters (grins, fangs, eye expressions).

Fashion and accessories (tattoos, collars, glasses).

Fantasy elements (animal ears, colorful aesthetics).