Chikasa style. Can do furries and humans. A bit chubby.
Trained using FurryToonMix as the base on 200+ pieces of artwork and 50 b&w manga crops for better pose and style understanding.
Put "grayscale, monochrome, manga" in negative prompt if you are getting black and white results.
No triggers.
V3:
You can use "facecrop" tag for eye detailer or face closeups to get (probably) better eyes and faces.
Added few characters:
Laika, blonde hair, black fur, countershading, blue eyes, dog, dog ears, dog tail
Ichigo, pink hair, pink fur, countershading, red eyes, brown gloves \(marking\), brown socks \(marking\), fox, fox ears, fox tail
Description
The better version. The last one had a low Max Token Length, which could result in parts of longer prompts being ignored or misunderstood.
Also this one is trained with regularization and caption learning.
FAQ
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.