Modified Anima to get rid of the quality tags(e.g. "masterpiece, best quality...", "worst quality, low quality...")
The quality tags are optional, but using certain meta tags and aesthetic tags can make your illustrations more beautiful.
The base model is Anima Base.
All the sample images were generated without the quality tags.
This model does not have the "AI Masterpiece" style.
This model supports artist tags.
日本語:
品質タグを不要にしたAnimaです(例: "masterpiece, best quality...", "worst quality, low quality...")
品質タグは任意ですが、一部のmetaタグやaestheticタグを使うことで、イラストをより美しくできます。
このモデルの画風はマージモデルに多い、いわゆる「AIマスピ絵」ではありません。
Noob版と同様に典型的なAIイラスト感を抑制しつつ美しいイラストを生成します。
ベースモデルはAnima Baseです。
すべてのサンプル画像は品質タグなしで生成しました。
アーティストタグ対応。
Text Encoderの仕様上、SDXL版のように勝手に美しいイラストを生成することはほとんどありません。足りない部分をDanbooruタグと自然言語(英文)で補うとよいでしょう。
Recommended settings/推奨設定
Steps: 14-32(fast: 14, balanced: 20, quality: 26)
Sampler: RES Multistep, ER SDE, Gradient Estimation, SA Solver
Scheduler: Simple, SGM Uniform, Normal, Linear Quadratic(requires >= 26steps)
Guidance Scale: 2-6(best value is 3.5)
Flow Shift(ModelSamplingAuraFlow): 2-4(best value is 2.5)
Tips: The amount of image error and detail is inversely proportional to the Flow Shift value.
VAE: Default(Qwen Image VAE) or Wan 2.1 VAE
Incompatible schedulers: Karras, Exponential, KL Optimal
Requirements
ComfyUI
VAE and Text Encoder not included. Please download it from here
Note/注意
version 0.1.0 does not support sd-scripts and diffusion-pipe.
version 0.1.0はsd-scriptsとdiffusion-pipeは非対応です。
Notice
This model provided "as is", without warranty of any kind.
This model is not designed for inpainting.
このモデルは「現状のまま」提供され、一切の保証はありません。
このモデルはinpainting用ではありません。
Description
Higher aesthetic score?
The base model is Anima Yume 0.4
FAQ
Comments (3)
That's the first time I comment on any model, but that's only because I just broke my personal record with this one, totalling 9 fingers per hand.
This model does not add weird artifacts like extra “buttons, pockets, cuffs” etc… to my display model like other finetunes, and the color is more vivid, brighter and preserve my original training image details. Recommended as showcase.
Very beautiful results. Still tends to do the "anime face" that has different coloring to the rest of the piece, so hopefully that can be fixed in future.




