v0.29: Distillation algorithm was almost completely rewritten.
Increased diversity. Different seeds now can generate more different images. This also improved lighting range, styles and LoRA compatibility.
Better details. This version can squeeze every single pixel out of the VAE.
See Update Log section for more version info.
See this page for standalone LoRA.
RDBT [Anima]
Distilled model (aka. "turbo"). It delivers faster speed and higher aesthetics with only 12 NFEs, 5x faster than base model (60 NFEs).
Sharing merges using this model is not allowed.
Usage:
Settings:
Sampler: "euler_a" "euler" "er_sde". (from smooth to high variance)
Steps: 8, 12 or 24. Scheduler: simple. Important: training timestamps are fixed. Other inference timestamps might not work.
CFG scale: 1~2. Cover images are all without CFG (CFG 1). You can enable CFG (CFG >1) if you need higher prompt adherence (e.g. style is too weak).
Prompt
Prefer natural language prompt. Prompt structure: style, subject, action, background.
Style is required! This model does not provide a default style. You should always prompt specific style. Or use a style LoRA. Otherwise, you will get random/mixed style.
There are some "rough" trigger words, they are trained but they aren't mapped to specific style:
"digital anime illustration": common 2d anime.
"digital art", 2d art but not anime, mostly digital art. (not many samples)
"anime sketch": simplified/unfinished anime drawing. (not many samples)
Quality tags:
You can omit all the quality tags. 1) The quality of training data is higher than "masterpiece". 2) Quality tags have been reinforced during distillation. Thus they don't have noticeable effects.
Same as negative tags. If you use cfg, there is no need to dump "score_1, blurry, worst quality, jpeg artifacts, extra arms,... x100 words" in your negative prompt. Those things have been distilled out.
Released models
RDBT LoRAs: Released as LoRA. For better distribution efficiency.
Update Logs
(4/23/2026) v0.27: Improved stability, details.
(4/18/2026) v0.25: It's based on anima p3.
Previous testing versions, see RDBT LoRAs