Update log for RDBT Anima
Links to base model:
AnimaYume (ym) (hf link) (civitai link).
Anima pretrained (b1,p3,p2,p1) (hf link)
(6/8/2026): b1 v0.39.b:
Same finetuned base, different distillation settings.
Higher diversity. More dynamic. Less Al sloppiness. (At the cost of stability, of course)
Tips: euler sampler has highest diversity, (estimated_) 2rd order samplers have lower diversity.
(6/4/2026): animayumev0.5 v0.39:
tldr: Comparing with precious v0.37. v0.39 don't have bias, thus has better style compatibility and prompt adherence. Also finetuned from scratch with 2x dataset and improved distillation settings.
In previous versions (probably from v0.24) after general finetuning, I also added a 2rd stage finetuning with highest aesthetic images. Sounds cool, but this literally equals to overfitted the model with smaller dataset, and introduced high bias (fixed pose, similar faces, expressions, shifted style, etc...) which is not ideal for a "base model".
So, v0.39 dropped this 2rd stage aesthetic finetuning. Because of this, v0.39 don't have built-in aesthetic trigger words.
Is this a downgrade? Noop, I would say it's a fix. You can grab any "aesthetic" LoRA you want. There are so many awesome LoRAs. Bias is good if it's exactly what you need, and is problematic if you don't want it. It's all about choice, and now you have it.
(5/24/2026): b1 v0.37.1:
Improve high noise stability a little bit.
(5/18/2026): b1 v0.35:
No step distillation. Just guidance distilled.
I'm dropping step distillation. Anima official has their plan to do step distillation (aka, turbo, 4/8-step, or whatever). They have the money and recourse and full dataset. I don't. And my cheap step distillation is kind of sh*tty, tbh.
If you need higher stability or speed, you can stack the extracted cosmos dmd2 lora the anima-turbo, basically can achieve the same thing, probably even better. I prefer 0.2x cosmos dmd2 lora.
(5/12/2026): p3 v0.32.b:
Less step distillation (means higher diversity but less stability). 12 steps is still doable, 24 steps is recommended for complicated prompt.
Styles reinforcement learning. I did this in v0.29, but not in v0.32.
(5/10/2026): p3 v0.32:
No more green-ish, color shifting.
Trigger words have been reclassified to avoid model learning a unified style. Update: those trigger words work in v0.37.1~v0.32:
@anime sketch: Low complexity. Rough outlines.
@digital anime illustration: Typical "anime". Clear and fine outlines. General complexity.
@digital art: More complex lighting, textures than typical "anime".
@cinematic digital art: More lighting, postprocess effects, semi-realistic, etc. A little bit chaotic.
Previous trigger words for backup (v0.29 and before):
"digital anime illustration": common 2d anime.
"digital art", 2d art but not anime, mostly digital art.
"anime sketch": simplified/unfinished anime drawing.
(4/29/2026) p3 v0.29.b: Further trained v0.29.
Known issues: v0.29 will add lots of details and make the images looks very sharp and clear. But this might affect other styles.
(4/27/2026): p3 v0.29: Distillation algorithm was almost completely rewritten.
Better details. This version can squeeze every single pixel out of the VAE.
(4/23/2026) p3 v0.27: Improved stability, details.
(4/18/2026) p3 v0.25: It's based on anima p3.
====== alpha testing ========
[base model version] [finetuned model version] [distillation method] [distillation version]
f = finetuned, cfg = cfg distilled, dmd2 = distribution matching distillation.
(4/14/2026) p3 v0.24f dmd2 b: Comparing with "v0.24f dmd2": Images should be sharper.
(4/11/2026) p3 v0.24f cfg b: Improved stability and diversity, comparing with previous 0.24f cfg.
(4/10/2026) p3 v0.24f dmd2: roughly 12~16 steps distillation. This is intentional. Low steps distillation = ai slop style without complex texture.
(4/8/2026) p3 v0.24f cfg: Rebased on preview3. Finetuned base model has -40% steps than v0.23. Less overfitted (?).
(4/4/2026) p2 v0.23f dmd2 b: Different distillation settings. Almost a 4 steps dmd2. Maximum stability.
(4/4/2026) p2 v0.23f dmd2: 8 steps dmd2. First dmd2 anima (?).
(3/28/2026) p2 v0.23f cfg: Rebased on preview2. Distillation: improved small details and stability (removed a regularization in distillation target and changed to second-order method).
Voting result: v0.20fd won. Thanks for the feedbacks.
(3/24/2026) preview1 v0.20f cfg b: Distillation: Different settings optimized for anime, high contrast and saturation.
(3/23/2026) p1 v0.20f cfg: Dataset: More furry. Finetuned base model: from v0.12 + more 100% steps. Distillation: Fixed noisy pixels this time, really.
(3/14/2026) preview v0.19f cfg b:
Updated dataset. Some private datasets have been dropped. You might notice the style changed.
Fixed high-freq artifacts in v0.12, now you should get a clear image without noisy pixels
b: Testing new distillation settings. Higher contrast. Aligned with common anime art.
(2/19/2026) preview v0.12f cfg:
Better stability and details, extended dataset.
(3/8/2026) preview v0.11f cfg 512px:
Prove of concept version for v0.12. Same dataset and settings as v0.12, except it was trained with 512px res.
Released by request, as it might be very useful. Running model in 512px and cfg1 is extremely fast (x10 faster, e.g. 30s -> 3s). If you don't have a beefy GPU. You can use this version to test your ideas/prompts in few seconds.
(2/12/2026) preview v0.6 cfg:
CFG distilled only. No finetuning. Cover images are using Animeyume v0.1.
(2/3/2026) preview v0.2f cfg:
Speedrun attempt, mainly for testing the training script. Limited training dataset. Only covered "1 person" images plus a little bit of "furry". But it works, and way better than what I expected.