Contains multiple characters from ????? Version. In my experience, the same lora containing multiple characters allows you to output them together in images more consistently.

In its current iteration, it should recognize the following:

trainerallie

allieoutfit

traineremmie

emmieoutfit

trainerjamie

jamieoutfit

trainermiranda

typhlosion

charizard

v1.0: This is my first attempt at training a model for Illustrious. It is based off of my own illustrations of the characters, though I hope tagging has made it agnostic in regards to style enough. It may be a bit rough in its first iteration, but I'll figure out the optimal settings.