[Illustrious] Shoomimi (Base Outfit) - Indie VTuber

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 Shoomimi in her base outfit. Trained on a dataset of only 26 images that largely consists of cut up screenshots enhanced by ChatGPT so it may not be accurate and have inconsistent results. This is especially in regards to small details like the seams and bolts on her body if you choose to prompt for those. The brain on top her head is also hit or miss but it works well enough. 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 and eyes. This LoRA was explicitly designed to work with an eye detailer as 4 of the 26 images in the dataset are extreme closeups of her eyes.

Manual inpainting or image refinement through ChatGPT are recommended if you're looking to generate seam lines and bolts.

May be updated some time in the future since I do want to improve consistency but the current V1 is pretty adequate for general use.

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

Triggers:

ShoomimiBase, short hair, hair loops, blue hair, blue eyes, facial scar, robot girl, medium breasts,

lab coat, red overalls, red skirt-like pants, wide-leg pants, blue pockets, white plus sign on chest pocket, pink plus sign on neck, flowy fabric, dark blue shirt, turtle neck, dark blue shoes, dark blue socks, short socks,

Experimental Triggers (Do not use if you want cleaner generation):

very small brain mascot in very small glass dome,

seam lines on arms, seam lines on legs, seam lines on ankles, bolts on skin, silver bolts,