Sign In

Remielle Dan (exp.) (Zenless Zone Zero) Anima

6

Download

1 variant available

fp32 SafeTensor

remielledanExpAnima.safetensors

Full precision, largest file • 63.16 MB

Verified:

Type

LoRA

Stats

106

6

13

Reviews

Published

Jul 9, 2026

Base Model

Anima

Hash

AutoV2
BD1C81FAD8
default creator card background decoration
Followers - 1069

1.1K

Likes - 6684

6.7K

Downloads - 41398

41.4K

License:

Anima

The Anima Model is licensed by CircleStone Labs LLC. Copyright CircleStone Labs LLC. IN NO EVENT SHALL CIRCLESTONE LABS LLC BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH USE OF THIS MODEL.

Built on NVIDIA Cosmos

00002-2570641229.png

Redistribution is prohibited.

If you like my work, consider giving me more support, likes, buzz, feedback and reactions.


ABOUT THE MODEL


Remielle Dan From Zenless Zone Zero

20260709v2

Trained on Anima Base v1

Finally managed to tune this style. It's now more close to the official artwork.

Zenless Zone Zero has an outstanding art style. However, its official artwork is heavily covered with film grain and various kinds of noise. Although these illustrations appear to be simple cel-shaded images to the human eye, for a model the size of Anima, those seemingly insignificant artifacts actually corrupt the underlying visual structure.

If the official artwork is used directly as training data, it results in severe information loss and distortion—film grain, JPEG artifacts, and other compression noise prevent the model from faithfully learning and reproducing the style.

In the end, I carefully curated the dataset and even created some of my own training images, making the style much more learnable for Anima while minimizing information loss during training.

Perhaps once a larger and more capable base model becomes available, this kind of complicated preprocessing won't be necessary anymore.

Features

  • 2+ outfits.

Issues

  • The costume details lack accuracy.

  • This model is designed for a specific art style, so the @ symbol is intentionally bound to its corresponding trigger word. If you'd like to blend it with other artist styles or experiment with different stylistic combinations, consider using a LoRA or tools such as Anima-Artist-Mixer to achieve the desired result.

  • When the face becomes blurry, consider using ADetailer to fix it.

  • When the output becomes blurry, consider using MultiDiffusion to fix it.

TRIGGER WORDS


Black

@remielledan,
pink hair, bob hair, side-swept bangs, black hair bows, silver-hairpins, a silver star-hairclip on hair, pink eyes, pointy ears, black gothic cyber dress with silver star ornaments, white high collar, sheer black sleeves, black long flared gloves, black wings, black lace thighhighs, black pointed heels with bows,

White

@remielledan,
pink hair,bob hair,side-swept bangs,large-violet-fabric-hairbow,two silver star-hairclips on hair,white headband,pink eyes, pointy ears,form-fitting white and silver strapless bodice,white wings,white detached flared sleeves,white stockings with a bow,silver-high-heels with a small decorative bow,

You could find the example prompts from my showcase.

My prompts are basically composed in the order of [character traits] + [clothing] + [camera]+ [customize content] + [background], and you can delete or modify them as needed.

Requirements


Checkpoints: Anima

Recommended Tools: ForgeNeo, ComfyUI

Recommended Step: 20~30

Recommended weight: 0.6~1

Recommended size: 768*1344, 832*1216, 896*1156, 1024*1024, 1344*768, 1216*8

32, 1156*896


I’ve set the gallery moderation of this post to “R”, so explicit content may not appear on this page.

ARCHIVES


20260704v1

Trained on Anima Base v1

Features

  • 1+ outfits.

Issues

  • Style is not stable enough, consider increasing the weight of the tag "a masterpiece image of 7em1e11e" or adding the tag "anime screenshot"

  • The costume details lack accuracy.

  • When the face becomes blurry, consider using ADetailer to fix it.

  • When the output becomes blurry, consider using MultiDiffusion to fix it.