Sign In

shaved ice / かき氷 / カキ氷 / 氷菓子

293

Download

1 variant available

bf16 SafeTensor

shavedice_anima_v1.safetensors

BF16, good balance • 87.61 MB

Verified:

Type
LoRA
Stats

222

109

118

Reviews
Published

Jul 19, 2026

Base Model

Anima

Usage Tips
Strength: 0.8
Hash
AutoV2
7AAA2F2BBE
Trigger Words
shaved ice
default creator card background decoration
Followers - 7362

7.4K

Likes - 79252

79.3K

Downloads - 572489

572.5K

CivChan Stabby Stabby Valentine 2026

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

The English description is translated below the Japanese one at DeepL.

屋台などで売っているカップに入ったカキ氷を食べるLoRAです。元からそれっぽい容器は出ますがほぼ完璧に再現できるようにします。

トリガーワードは shaved ice です。

重みは0.8前後推奨。

トリガーだけだと大体持つだけになるので食べるのであれば holding drinking straw, spoon straw, eating 辺りを入れて下さい。

Anima版はAnima_base Version1をベースとして学習していますので他のAnimaマージモデルでも使用できる可能性はありますが検証は行っておりません。

Illustrious版はIllustrious-XL 0.1をベースとして学習していますので他のIllustriousマージモデルやNoobマージモデルでも使用できる可能性はありますが検証は行っておりません。

Animagine版はAnimagine XL 3.1をベースとして学習していますが3.0でも動作は確認しています。それ以外のXLモデルでは正常に機能しない可能性があります。

Pony版はPony Diffusion V6 XLをベースとして学習していますので派生モデルでは正常に機能しない可能性があります。

以下翻訳

This is a LoRA model for eating shaved ice served in cups, like the kind sold at food stalls. While the model already generates containers that look similar, we’ll fine-tune it to achieve a nearly perfect recreation.

The trigger word is “shaved ice.”

We recommend a weight of around 0.8.

If you use only the trigger word, the model will usually just hold the cup; to make the character eat, please include words like “holding,” “drinking straw,” “spoon,” “straw,” or “eating.”

The Anima version was trained using Anima_base Version 1 as a base, so it may work with other Anima merge models, but this has not been verified.

The Illustrious version was trained using Illustrious-XL 0.1 as a base, so it may work with other Illustrious merge models or Noob merge models, but this has not been verified.

The Animagine version was trained using Animagine XL 3.1 as a base, but we have confirmed that it works with version 3.0 as well. It may not function properly with other XL models.

The Pony version was trained using Pony Diffusion V6 XL as a base, so it may not function properly with derivative models.