<h1 id="2026-anima-model"><strong>⬇️2026 Anima model⬇️</strong></h1><h3 id="the-anima-model-is-trained-using-the-same-training-image-data-as-the-illustrious-version-(though-all-tags-have-been-reconfigured)."><strong>⚠️</strong><span style="color:rgb(250, 176, 5)"><strong>The Anima model is trained using the same training image data as the Illustrious version (though all tags have been reconfigured).</strong></span></h3><h3 id="all-training-materials-consist-of-landscape-oriented-images.-we-strongly-recommend-specifying-a-landscape-aspect-ratio-when-generating-images."><span style="color:rgb(253, 126, 20)"><strong>⚡All training materials consist of landscape-oriented images. We strongly recommend specifying a landscape aspect ratio when generating images.</strong></span></h3><h3 id="recommended-strength-for-the-lora-is-between-0.85-1."><strong>⚠️Recommended strength for the LoRA is between 0.85 - 1.</strong></h3><h3 id="(landscape)width:1216-height:812-is-best"><strong>🖼️(LandScape)Width:1216 Height:812 is best</strong></h3><hr /><h3 id="you-can-effectively-switch-between-anime-and-realistic-styles-by-specifying-negative-prompts."><span style="color:rgb(250, 82, 82)"><strong>📢You can effectively switch between anime and realistic styles by specifying negative prompts.</strong></span></h3><h3 id="negative-prompt-(anime-style)"><span style="color:rgb(250, 82, 82)"><strong>✒️Negative Prompt (Anime Style)</strong></span></h3><pre><code>(realistic:1.5), photorealistic, </code></pre><p><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/dff027d9-75f3-4f6f-8c25-8d59639c3bc0/original=true/dff027d9-75f3-4f6f-8c25-8d59639c3bc0.jpeg" /></p><hr /><h3 id="if-no-negative-prompt-is-specified"><span style="color:rgb(250, 82, 82)"><strong>✒️If no negative prompt is specified</strong></span>　</h3><p><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/0bf88090-8655-4df7-909c-d72738bb9a58/original=true/0bf88090-8655-4df7-909c-d72738bb9a58.jpeg" /></p><p>Selecting "Realistic" will result in a realistic look, but I do not recommend it.</p><hr /><hr /><h1 id="2025-il-model"><strong>⬇️2025 IL model⬇️</strong></h1><h1 id="this-lora-was-created-to-see-if-it-could-learn-jav-and-produce-hentai-anime."><span style="color:rgb(250, 176, 5)">This LORA was created to see if it could learn JAV and produce hentai anime.</span></h1><h2 id="i-think-lora-strength-of-0.5-to-0.7-is-appropriate-for-the-il-version."><span style="color:rgb(18, 184, 134)"><strong>I think LORA strength of 0.5 to 0.7 is appropriate for the IL version.</strong></span></h2><h3 id="width:832-height:1216-is-better"><strong>Width:832 Height:1216 is better</strong></h3><h3 id="width:1216-height:832-is-best!!"><strong><u>Width:1216 Height:832 is best!!</u></strong></h3><p><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/a7831f5c-7df0-4932-b3b9-a4ccbb7a0ff5/width=525/a7831f5c-7df0-4932-b3b9-a4ccbb7a0ff5.jpeg" /><span style="color:rgb(250, 82, 82)">learned Real images. If you are creating hentai anime, set the following Negative:</span></p><p><span style="color:rgb(250, 82, 82)"><strong>ADD Negative (HENTAI ANIME)</strong>---------------------------------------------------------------------</span></p><h3 id="(3d-photorealistic-lifelike-realistic-true-to-life-vivid-picture-perfect-naturalistic-convincing)"><span style="color:rgb(250, 82, 82)">(3D, Photorealistic, Lifelike, Realistic, True to Life, Vivid, Picture-Perfect, Naturalistic, Convincing), </span>　</h3><p>　</p><p><strong>SAMPLE </strong><span style="color:rgb(193, 194, 197)">------------------------------------------------------------------------------------------------</span></p><h3 id="deepthroat-irrumatio"><span style="color:rgb(18, 184, 134)"><strong>DeepThroat / Irrumatio</strong></span></h3><pre><code>&lt;lora:JAV_MIAA572_IL_V1:0.7&gt;
JAV_MISS572,  1girl, hetero, solo focus, tears, crying, saliva, snot, drooling,  irrumatio, deepthroat ,
1boy, male pubic hair, Fat man, nude male, Grab head,


&lt;***Clothed Character Prompt***&gt;</code></pre><p>　</p><h3 id="option(bukkake)"><strong>Option(Bukkake)</strong></h3><pre><code>bukkake, facial, excessive cum ,  close-up mouth , vomiting</code></pre><h2>　</h2><p>　</p><p><strong>--sorry Japanese text-- -----------------------------------------------------------------</strong></p><p><span style="color:rgb(250, 176, 5)">このLORAは、</span></p><h2 id="av"><span style="color:rgb(250, 176, 5)">「実写AVを学習して、内容をアニメの任意キャラで生成できるか？」</span></h2><p><span style="color:rgb(250, 176, 5)">を検証するために作成されました。（その⓶）</span></p><p><span style="color:rgb(250, 176, 5)">概ね良好な結果となりましたので、公開するに至りました。</span></p><p></p><h1><span style="color:rgb(34, 139, 230)">横長でのランダム性の高い生成を推奨します。縦長は単調な生成となりがちです。</span></h1><p>　</p><ul><li><p>サンプルスクリプトの　&lt;***Clothed Character Prompt***&gt;　部分にお好きなキャラクターを指定してご使用ください。</p></li><li><p>実写画像を学習しています。<span style="color:rgb(250, 82, 82)">アニメ絵を生成する場合は、Negativeに　(3D, Photorealistic, Lifelike, Realistic, True to Life, Vivid, Picture-Perfect, Naturalistic, Convincing), </span>　を指定してください。</p></li><li><p>IL版でアニメ出力する場合は、LORAの強度は0.6～0.7程度がいいと思います。</p></li><li><p>LORAの強度を上げるとリアル寄りに変化します。ご注意ください。</p></li><li><p>キャラクターLORAと併用する場合は、キャラクター側LORAの強度をかなり落とすことをお勧めします。</p></li><li><p>他のリアル系LORAと併用した場合、相乗効果で写真描写が強調される現象が発生します。ご注意ください。</p></li></ul><p></p><p>サンプルスクリプトに限らず、何か面白いプロンプトやアイデアが見つかりましたら画像投稿していただけると、みんなが（主に私が）幸せになります。（できれば生成プロンプト情報は削除しないでください♥）</p><p></p><hr /><p>　</p><p>The AI ​​model creation support software "<a target="_blank" rel="ugc" href="https://civarchive.com/articles/24674/create-models-from-videos-optimized-software-has-been-released">HENTAI APP</a>" is used to create the models.</p><p><img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/85b4dbd3-9847-4cb8-bdf1-d6e3dd91e817/width=525/85b4dbd3-9847-4cb8-bdf1-d6e3dd91e817.jpeg" />Please refer to the article below for details<a target="_blank" rel="ugc" href="https://civarchive.com/articles/24674/create-models-from-videos-optimized-software-has-been-released">：</a><a target="_blank" rel="ugc" href="https://civarchive.com/articles/24674/create-models-from-videos-optimized-software-has-been-released">https://civarchive.com/articles/24674/create-models-from-videos-optimized-software-has-been-released</a></p><hr /><hr /><p>　　　</p><p><span style="color:rgb(193, 194, 197)">私は本LORAの利用に、制限・権利・責任を一切主張しません。</span></p>

Files with SHA256: ea5940b270f4a41e529e81f1d384b34d48b5f1c0fbd1babc6e4ba423638591e9