Change Browsing Level to "R" if you want to see the full set of example images.

A very simple character LoRA for the indie VTuber Sopralto in her villainess model/outfit. Trained on a dataset of 40 images that largely consists of cut up screenshots enhanced by ChatGPT. Although the dataset is on the larger side, smaller details are still hit or miss. The heart-shaped cutout sometimes has issues generating properly and the flower lining on the breast cups can be difficult too. 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. Increasing clip strength can generally yield more consistent results.

Detailers are highly recommended for the face and eyes.

Manual inpainting or image refinement through ChatGPT are recommended if you're looking to generate small details.

Probably won't be updated.

All example images were generated without a refiner step. You can probably make your generations look much better with improved workflows.

Triggers:

SopraltoBig, long hair, single braid, large braid, ahoge, heart-shaped ahoge, blonde hair, blue eyes, guinea pig girl, guinea pig ears, large breasts,

golden crown, neck ribbon, purple jacket, large jacket, fluffy jacket, white leotard, heart-shape hole in leotard lined in gold, single gold thigh ring, single ankle ribbon, barefoot,