(22/10/2026): Next: MOSS Image (aka. The greatest anime model that has ever been released.). Here.
RDBT [Anima]
A finetuned model from a personal all-in-one dataset, mainly for digital art and drawings, focusing on lighting and textures, from wallpaper level digital art to 5-min sketch.
Dataset contains ~10k images, handpicked by professional amateur artists. The criteria for evaluation are not visual appeal, but the drawing technique and the level of completion. Captioned by LLM.
Some cover images look ridiculous; they're just for demonstration purposes for corner cases.
Update log and version info: link.
Useful links:
RDBT | Digital Art: My merged 16-step ckpt optimized for digital anime art. 4x faster. Easy to prompt.
RDBT | Original LoRA: LoRA files for advanced users.
16-step Acc LoRA: Style friendly, less AI slop, compared to the official Turbo.
RDBT | 160KE6 Furry: Based on 160kE6Anima by @garaggge. 160kE6Anima "was trained on around 160k images from e621". It has extended knowledge and 3k+ style tags. Check its model page above for more info (list of styles, etc.).
Settings:
Base version: CFG 4 and steps 24+. Euler.
Discontinued:
Distilled version (8s, 16s, discontinued): CFG: 1. Steps: 8 or 16. Euler.
Prompt:
Always specify @style in prompt, or use a style LoRA. This model does NOT provide overfitted default style (ignore prompt and is always active). You need to specify style in prompt, or use a style LoRA. This is a feature, not a bug. The dataset is big enough and model is not overfitted. Otherwise, you will get random/mixed style.
Quality tags:
You don't need quality tags. The dataset has higher quality than "masterpiece". Thus quality tags don't have effects. Omitting those redundant tokens allows LLM to pay more attention on other words.
Sharing merges using this model is not allowed.
This mode is a free and it will always be free. This "restriction" won't affect anyone. It's only aimed at those who steal others' models to sell.
Known model thieves: NukeA.I (selling this model behind paywall on tensorart).
Because he is such a jerk I wrote a story about it. Also contains a guide for trainers about "how to bake special trigger word into your model".
Description
FAQ
Comments (14)
Looks great. Really good work well done 👍. Have a ever considered collaborating with @duongve13112002 to make a solid, stable and high quality Preview 2 checkpoint? As I love preview 1 but the natural language prompt adherence is better especially for more complex scenes and with multiple subjects on P2.
Any examples? I tested p2 and I think prompt adherence is the same as p1, afaik, they were still using tag based captions. And RDBT has better prompt adherence than p2, at cost of built-in styles.
the original p1 does have some stability issues, but it is a "pretrained" model, it should be unbiased and thus unstable. It's ideal for finetuning, but not ideal for users.
I think p2 is a finetuned model, which is difficult to finetune again. Some people said it takes longer to train LoRA on p2, that might be the prove.
@reakaakasky it's been a while since I used it but what is your opinion on NetaYume Lumina aa I think it's great I don't know how it compares to anima tho.
Lumina has way better aesthetics than Cosmos predict 2.
Cosmos predict 2 has better logical understanding.
How are you training these ? LCM ?
just cfg distillation
I see , thank you
Was just curious to try as well.
Tried LCM but it didn't turn out that well , it may need same dataset as original model was trained on.
Works quite well with https://civitai.com/models/2466415/cosmos-predict25-2b-base-distilled-extracted-dmd2-lora at 0.7 strength, 12 steps cfg 1, and https://github.com/pamparamm/ComfyUI-ppm for somewhat working negatives
(silly example prompt: "2girls, kissing (score_9, blushing, :-1.0)" )
very cool
Do you have an fp8 version of anima preview 2?
no, I gave up. it's slower in ComfyUI. fp8 needs torch.compile to inline kernels. Right now torch.compile is unusable, and will be unusable forever if they enforce their dynamic vram mode. On my hardware it's even slower 20% than bf16 +compile.
https://huggingface.co/Bedovyy/Anima-FP8/tree/main
Not mine, not tested, enabled hw fp8, but no calibration metadata.
@reakaakasky Yea, i'm rolling now with silveroxides int8 quant. Slightly faster than fp8 while having basically bf16 quality. 2.34s/it vs. 1.58s/it on my hardware fp16 preview2 vs int8 preview2, no torch compile. Seeds look fairly similar. Btw you were right about klein anime finetune in the making. Apparently chenkin is at it.
. https://huggingface.co/silveroxides/Anima-Quantized/tree/main
@deitychaser I don's see the chenkin klein finetune on their page, or did they just start?
@mc355168 Yea, its still in trainig in testing.
Man all of my lora floading here since your checkpoint is my favorite and use it to generate sample for them lol Still enjoying the checkpoint very much thankfull for it !
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.









