LoRA for generating in the style of doujinshi artist Shrimp Cake.

MOSS

Trained on MOSS-Image-Base, tested on MOSS-Image-SFT

Examples made on default comfy workflow with hires fix.

er_sde + simple, 25 steps and 8 hires turbo, cfg 3, lora strength 1. Resolution - 2 mpx on multiple of 16.

Stating the obvious - to render in this style you need it to be illustration, 2d, anime, etc.

Tl:dr do not use it with realistic description if your goal is to emulate this style

Anima

Dataminers for meta tags shows that Anima knows this artist, but it's not it, probably poor generated ai data on booru.

Tested on turbo, base, and aesthetic (not recommended).

Turbo on base lora gives more close resemblance to the original, but may cause additional effects like steam and sweat.

Use it without turbo on not official checkpoints.

Tested on natural language, CLIP styled prompts and mix of them.

All of examples are made on 1 strength and cfg 1-1.5.

Quality tags masterpiece, best quality, score_7 and such are important.

Workflow is inside of the pictures.