Vivziepop Style

Trained as an experiment using a single source image with Perceptual LoRA Toolkit. Seems pretty flexible and adheres well to prompts. Happy to hear feedback about how to improve it.

Note that this is not trained with character names, just physical descriptions.