中文说明

Uto（ウト）是 Eve《風のアンセム feat. suis from ヨルシカ》MV 中登场的角色，角色设计与动画由くっか负责。

这是一个基于 Anima 训练的非官方角色 LoRA，主要用于还原 Uto 的人物形象，包括红棕色短发、暖棕色眼睛、带暖象牙白发梢的宽大侧发束，以及白色褶边兔耳形布艺头饰。

白色兔耳形结构是角色佩戴的布艺头饰，不是生物兔耳。

模型信息

• 类型：Character LoRA

• 基础模型：Anima

• 训练图片：48 张

• 正则图片：48 张

• 训练轮数：30 epochs

• 总训练步数：720 steps

• LoRA Rank / Alpha：32 / 32

• 推荐权重：0.8–1.0

建议从 0.8 开始测试。其他角色 LoRA 或较强的画风 LoRA 可能会影响人物还原效果。

基础人物提示词：

1girl, uto, maid headdress, brown eyes, solo, portrait, looking at viewer, simple background

换装时建议保留：

uto, maid headdress, brown eyes

其中 maid headdress 用于加强 Uto 固定的白色褶边兔耳形布艺头饰。

标志性女仆装

uto_maid_outfit

推荐提示词：

uto_maid_outfit, uto, 1girl, solo, full body, black dress, white frills, black boots, standing

进行自由换装时，请不要使用 uto_maid_outfit。

使用注意

• - 本模型仅训练了 Uto 的人形形态。

• - 红伞和苹果形扣饰需要在提示词中单独写出。

• - 换装时建议继续保留 maid headdress 和 brown eyes。

• - 不建议在负面提示词中加入 animal ears 或 rabbit ears，否则可能压掉白色兔耳形布艺头饰。

• - 测试模型时建议关闭其他角色 LoRA。

本模型为非官方粉丝制作的角色 LoRA。角色、作品及相关视觉设计的权利归原作者和相关权利方所有。本模型与官方创作者无关联，也未获得官方授权。

English Description

Uto is a character featured in Eve's “Kaze no Anthem feat. suis from Yorushika” music video. The character design and animation were created by Kukka.

This is an unofficial Character LoRA trained for Anima. It focuses on reproducing Uto's auburn short hair, warm brown eyes, broad side hair sections with warm ivory tips, and distinctive white ruffled rabbit-ear-shaped fabric headpiece.

The rabbit-ear-shaped structure is a fabric headpiece worn by the character, not a biological animal feature.

Model Information

• - Type: Character LoRA

• - Base model: Anima

• - Training images: 48

• - Regularization images: 48

• - Training: 30 epochs / 720 steps

• - LoRA Rank / Alpha: 32 / 32

• - Recommended strength: 0.8–1.0

Start at 0.8 and adjust when necessary. Other character LoRAs or strong style LoRAs may reduce character accuracy.

Basic identity prompt:

1girl, uto, maid headdress, brown eyes, solo, portrait, looking at viewer, simple background

For outfit changes, keep:

uto, maid headdress, brown eyes

The maid headdress tag helps preserve Uto's white ruffled rabbit-ear-shaped fabric headpiece.

Signature Maid Outfit

uto_maid_outfit

Recommended prompt:

uto_maid_outfit, uto, 1girl, solo, full body, black dress, white frills, black boots, standing

Remove uto_maid_outfit when changing her clothing.

Usage Notes

• - Only Uto's humanoid form was included in the training set.

• - Add red umbrella or apple ornament when those details are needed.

• - Keep maid headdress and brown eyes when testing outfit changes.

• - Do not add animal ears or rabbit ears to the negative prompt. These terms may suppress the white rabbit-ear-shaped fabric headpiece.

• - Disable other character LoRAs when evaluating character accuracy.

This is an unofficial fan-made model. Uto, the original work, and the related character design belong to their respective rights holders. This model is not affiliated with or endorsed by the official creators.

免责声明

本 LoRA 为非官方同人模型，与原作者、画师、开发商、发行方及版权方均无任何官方关联。

模型涉及的角色形象、原作设定、美术素材及相关知识产权均归各自权利方所有。上传者仅进行了个人兴趣性质的模型训练与分享，不主张对原角色或原作品拥有任何权利。

请勿使用本模型冒充官方内容、进行虚假署名、误导性宣传，或从事可能损害原作者及相关权利方权益的行为。

如原作者、版权方或其他相关权利人对本模型的发布存在异议，可以联系我，我会认真处理相关问题。

This is a fan-made and unofficial LoRA model, mainly created for AI image generation experiments, learning, and fan works.

I do not own the original character, artwork, source material, or any related intellectual property. All rights belong to their respective owners. This model is not affiliated with, endorsed by, or officially connected to the original creators, artists, publishers, developers, or copyright holders.

Please do not use this LoRA to impersonate official artwork, falsely claim authorship, mislead others, or infringe upon the rights of the original creators.

This model is mainly trained for character likeness. Because of the training dataset, certain outfits, poses, expressions, or background elements may appear more frequently than expected. If the character looks too fixed or the LoRA has too much influence on the image, try lowering the LoRA weight.

I am still learning LoRA training, so the model may not be perfect. Feedback, suggestions, testing results, and example images are always welcome.