PixAI会員
PixAI
PixAI
Generate
Sign in
ホーム
生成
モデル
Studio
NEW
コンテスト
Mio.2 で作成
ツールボックス
NEW
ガイド

ユーザーLoRA

XL
スタイル

SDXL Nose hook Anime style

#鼻フック
AIツール
  • AIアニメ生成
  • アニメフィギュア生成
  • キャラクターシート生成
  • 雑誌表紙生成
  • ぬいぐるみ生成
  • 巨大彫像生成

探索

  • 人気タグ
  • ランキング
  • モデルマーケット
  • コンテスト
  • お知らせ

PixAIについて

  • ガイド
  • PixAI ブログ
  • Tsubaki.2
  • マスコットのミオに会う
  • コンテンツ規則

料金とヘルプ

  • 会員
  • クレジットパック
  • お問い合わせ

アプリ

  • PixAI アプリを入手
  • App Store
  • Google Play
© 2026 PixAI
  • 利用規約
  • プライバシーポリシー
  • 著作権ポリシー
  • クレジット表記
  • 特定商取引法
SDXL Nose hook Anime style - AI Modelのカバー画像
アップロード日時
2025/01/22 13:40
採用数
1.9k
レビュー
良好  (1)
トリガーワード

nosehook ,One Girl,(Masterpiece, Top Quality, Very Aesthetic, Absurd, Super Detail, Hi-Res, 8K) <lora:SDXL_nosehook_anime_v5-000007:0.8>Nose Hook,Front

権限
  • 画像生成と共有を許可する
  • ユーザーによるモデルのダウンロードを許可する
  • 生成画像の商用利用を許可する

詳細説明

[new]V5.0 has been released.The influence of previous anime style drawings has been reduced.If you enter [anime] in the prompt, you will get the traditional version.I haven't tried it, but it may be that the pattern has less of an effect when crossing with other Loras.The learning file has been modified so that the nose is depicted slightly larger.The lips are also now pulled upwards. I haven't experimented with creating many images yet, so there may be many bugs.It supports faces from the front and at a slight angle.Triger word is [nosehook]3way nose hook Ver.1 has been released.I have set a trigger word for this [3waynosehook]Front and side views are shown. Most of the time, it doesn't draw well, probably due to a lack of learning.The problem is that if you turn to the side, your ears won't be covered.I'm thinking of upgrading when the environment is ready.Please try changing the strength between 0.7 ~ 1.3The best distance for me is around 0.8.The farther the subject is, the less accurate it will be.I made it for SDXL. It was made by an amateur. I don't know if it will work well.The file size is large.Some versions require triggers and some do not. If it is not reflected, please check carefully.The learning model is[Counterfeitxl_v25]<-v1.0[animagineXLV3_v30]<-v3.0 and 3way v1.0[animazinexlv3.1]<-v4.0I haven't been able to use it well myself.If you create a good work, please upload it here.

コメント