v1.0: I think it's time to release v1.0. I've migrated to the krea2 model (never imagined that krea2 would be my next choice...). Fine-tuning the anima model is really fun.I also uploaded int8 version for ComfyUI hardware int8 (labeled as fp8).8-step model has highest stability. 16-step has higher diversity.<hr /><h2 id="rdbt-anima">RDBT [Anima]</h2>This is a finetuned model with 10k high aesthetic images paired with natural language captions from LLM. Then distilled to further improve quality and stability. Dataset does not contain any shiny plastic glossy AI image.It's not overfitted and doesn't have a default style. I use it as a clean starting point to stack more style LoRAs. I can stack whatever I want and get exactly what I stacked.See <a target="_blank" rel="ugc" href="https://civarchive.com/articles/30582">this page</a> for update log.For advanced users: The RDBT model is trained as LoRA natively. See <a target="_blank" rel="ugc" href="https://civarchive.com/models/2364703">this page</a> for original LoRA, update more frequently.This model is based on:<ul><li>prefix with ym: AnimaYume (<a target="_blank" rel="ugc" href="https://huggingface.co/duongve/AnimaYume">hf link</a>) (<a target="_blank" rel="ugc" href="https://civarchive.com/models/2385278">civitai link</a>).</li><li>prefix with b,p: Anima pretrained (<a target="_blank" rel="ugc" href="https://huggingface.co/circlestone-labs/Anima">hf link</a>)</li></ul><hr />Sharing merges using this model is not allowed. If someone is selling this model as their own, I'm happy to list them here so everyone knows.Known model thieves: <a target="_blank" rel="ugc" href="https://tensorhub.art/u/613019935106639991">NukeA.I</a> (behind paywall on tensorart).I wrote a <a target="_blank" rel="ugc" href="https://civarchive.com/articles/31410/dev-story-rdbt-2">story</a> about it. Also contains a guide for trainers about "how to bake special trigger word into your model".<hr /><h3 id="usage:">Usage:</h3>Settings:CFG scale: 1~3. This model has been distilled. You can disable CFG (CFG 1) and run the model 2x faster. Cover images are without CFG for demonstration.Steps: 16+Prompt:Always specify style, or use a style LoRA. Otherwise, you will get random/mixed style. This is a feature, not a bug. This model does not provide a overfitted default style.Quality tags:It's recommended to omit all the quality tags, or just keep the "masterpiece", if you're not confident. Omitting those redundant tokens allows LLM to pay more attention on other words.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" into your negative prompt. Those things have been distilled out.<hr /><h3 id="training-settings">Training settings</h3>All captions are NL from Google Gemini.Optimizer: adamw, constant lr 0.00002, weight decay 0.1, batch size 16.LoRA rank/alpha 24.Timesteps shift 3.Block 0-2 and adaln linear layers are skipped.

RDBT | Anima

rdbtAnima_b1V108Step.safetensors

b1 v1.0 8-step

b1 v1.0 16-step

ym1f v0.43 [int8]

ym1f v0.43

b1 v0.39.b

ym0.5 v0.39

b1 v0.37.1

b1 v0.35

b1 v0.32.b

p3 v0.32

p3 v0.29

p3 v0.27

p3 v0.25

v0.25

p3 v0.24f 4-step dmd2

p3 v0.24f dmd2 turbo

p3 v0.24f 16-step dmd2

p1 v0.20f cfg

p1 v0.19f cfg b

p1 v0.12f cfg

fp16 patch v1.2

fp16 patch

p1 fp8

Files with SHA256: 0b302de67024459628335fe1a0c882884a7f43f9a657f85e5c1bbc4a6ad9a410