Change Browsing Level to "R" if you want to see the full set of example images.
A character LoRA for the FlaVR VTuber Minna Vanilla. Trained on a dataset of 40 images that consists almost entirely of images generated by ChatGPT due to the fact that no other material exists. Everything from the waist down is original due to the fact that Minna Vanilla is a character originally generated using NovelAI by Pipkin Pippa as part of a joke and does not actually have an official full-body design. Regardless, I think the full body design ChatGPT and I came up with is quite nice and continues the flow and feel of the original adequately enough. That being said, due to a the majority of the images in the dataset being purely AI generated, this model can have the tendency to be unstable and inconsistent. Increased clip strength is recommended. Bottom line: This model is experimental and not intended to the highest quality thing in the world. The below and attached triggers are general suggestions. You can almost certainly get better and more consistent results by more specific prompting. I'm just lazy. Play around with it to see what works for you.
Detailers are highly recommended for the face, eyes, and hair.
Manual inpainting or image refinement through ChatGPT are recommended if you're looking to generate small details.
May be updated in the future as I would like to improve consistency.
All example images were generated without a refiner step. You can probably make your generations look much better with improved workflows.
Triggers:
MinnaVanilla, short hair, ahoge, heart-shaped ahoge, silver hair, silver eyes, medium breasts,
blue hair bow, blue neck ribbon, bare shoulders, white collar, white dress, frilly dress, blue and white dress frills, blue waistband, blue waist ribbons, detached sleeves, white sleeves, long sleeves, blue thighstrap, white thighhighs, white frilly thighhighs, white frills, blue heels, platform Mary Janes, ankle-strap,