v4
Compatible with ComfyUI and Forge Neo.
For prompting guidelines, please refer to the official Anima model page.
Positive Prompt:
masterpiece, best quality, [safe / sensitive / explicit, uncensored]
Choose one rating tag: safe, sensitive, or explicit.
Negative Prompt:
worst quality, low quality, score_1, score_2, score_3, artist name
Recommended Settings:
Sampling method: Res Multistep
Schedule type: Simple
Sampling steps: 21–24 (22 recommended)
CFG Scale: 2.5–3.5 (3.0 recommended)
Resolution: Up to 1536 × 1536, depending on available VRAM,
Recommended Resolutions: 1024 × 1024, 832 × 1216, 896 × 1152, 1024 × 1344
Low-VRAM Option:
If you have limited VRAM, you may use the community-made INT8 Qwen3 0.6B text encoder. It can reduce memory usage, although it may not improve generation speed. The official text encoder is recommended if you have sufficient VRAM.
Description
Merged AnimaIka, animaCatTower, AnimaYume, and several other LoRAs.
This has reduced the variety of low-quality seeds.
FAQ
Comments (6)
What a nice model!
いいモデルですね！
Please can you release the Ikamusume lora you used? It seems unavailable for download on Civitai
Love you! (づ｡◕‿‿◕｡)づ
This model has two advantages:
1.The original anima reproduces the source material too faithfully. As a result, if the LoRA is trained from anime screenshots, the generated skin tones in anima often turn out flat. This model avoids that issue.
2.The original anima backgrounds look too plain, almost like clumsy coloring, so adding a photo background becomes necessary. But that makes the background overly realistic. With this model, even when a photo background is added, the background looks better without becoming excessively realistic.
The only subjective issue is that the colors are very intense, especially when lighting-related keywords are used.
I look forward to the author recreating one based on the anima base.
Are you talking about adding photo backgrounds via ControlNet or just the tag "photo background" itself in the positive prompt?
@AceAtLarge Photo background (prompt). You can try it with the original model and this model. If you call a person as the premise, you’ll notice that in the original model the background becomes realistic, as if the image were crudely post‑processed, placing the person into the picture.