Alpha V17 for Mugen

017.0/Mugen-1024/aidl-v017-mugen-1024.cm-info.json

@@ -0,0 +1,41 @@
+{
+  "ModelName": "[Mugen] AIDL: OmoAIshi",
+  "ModelDescription": "OmoAIshi is the most comprehensive multi-concept LyCORIS for diaper wearing and using, able to comprehend multiple concepts previously unfulfilled by base models and other LoRAs. LyCORIS LoCon is used instead of the regular LoRA format as it allows multiple concepts to be trained within the same model. No more girls in panties! \n\nThis LoRA violates CivitAI\\u0027s terms of service (their TOS does not allow diapers nor scat). Do not upload generations from AIDL onto CivitAI.",
+  "Nsfw": true,
+  "Tags": [
+    "diaper",
+    "abdl",
+    "omorashi",
+    "scat"
+  ],
+  "ModelType": "LoCon",
+  "VersionId": 0,
+  "VersionName": "v017",
+  "VersionDescription": null,
+  "BaseModel": "Other",
+  "FileMetadata": {
+    "fp": null,
+    "size": null,
+    "format": "SafeTensor"
+  },
+  "ImportedAt": "2024-12-12T07:27:13.2830617+00:00",
+  "Hashes": {
+    "SHA256": "thisisnotahash",
+    "CRC32": "thisisnotahash",
+    "BLAKE3": "thisisnotahash"
+  },
+  "TrainedWords": [
+    "diaper",
+    "peed diaper",
+    "poopy diaper",
+    "pull-ups",
+    "peed pull-ups",
+    "pooping self",
+    "poopy underwear",
+    "poopy shorts",
+    "poopy pants",
+    "open diaper"
+  ],
+  "UserTitle": null,
+  "ThumbnailImageUrl": "/opt/stabilitymatrix/Data/Models/Lora/OWN/aidl-v015-1-illustrious2-1536.preview.png"
+}

017.0/Mugen-1024/aidl-v017-mugen-1024.metadata.json

@@ -0,0 +1,62 @@
+{
+  "file_name": "aidl-v017-mugen-1024",
+  "model_name": "[Mugen] AIDL: OmoAIshi V017",
+  "file_path": "/home/pops/AI/ComfyUI/models/loras/Own/aidl/aidl-v017-noobai-rf-1024.safetensors",
+  "size": 509995232,
+  "modified": 1767142313.522548,
+  "sha256": "cbb06d67f38a4ee65bc20523f3514ddd8325c444a3424a2f2cebbe1811b70972",
+  "base_model": "Other",
+  "preview_url": "aidl-v017-mugen-1024.preview.png",
+  "preview_nsfw_level": 16,
+  "notes": "",
+  "from_civitai": false,
+  "civitai": {
+    "name": "V17 - Mugen",
+    "description": "<style>  .main-div {    background-color: #9999ff; padding: 20px; border-radius: 10px; margin-bottom: 20px;  }  h1 {    color: #24292e;    margin-top: 0;    font-size: 25px;  }  h2 {    color: #24292e;    margin-top: 0;    font-size: 22px;  }  h3 {    color: #24292e;    margin-top: 0;    font-size: 20px;  }  h4 {    color: #24292e;    margin-top: 0;    font-size: 15px;  }  p {    color:#24292e;  }  a {    color: #0366d6;    text-decoration: none;  }  .section-div {    background-color: white;    padding: 15px;    border-radius: 8px;    margin: 15px 0;    box-shadow: 0 2px 4px rgba(0,0,0,0.1);  }  .subsection-div {    background-color: #bbb;    padding: 15px;    border-radius: 8px;    margin: 15px 0;    box-shadow: 0 2px 4px rgba(0,0,0,0.1);  }  ul {    margin-bottom: 0;    color: #24292e;  }  ol {    margin-bottom: 0;    color: #24292e;  }  td {    margin-bottom: 0;    color: #24292e;  }  code {    background-color: #333;    padding: 3px 6px;    border-radius: 4px;  }  table {    color: #24292e;  }</style><div class=\"main-div\">    <h1>AIDL: OmoAIShi Changelog</h1>    <div class=\"section-div\">        <h2>V16: Tag Restructure Edition</h2>        <p>Welcome to V16! Refined tags, another new customizable diaper element, and the new Neta Lumina architecture brings AIDL to the world of Diffusion Transformers. While the Lumina variant is the standout variant of AIDL, the SDXL variants can still hold its weight well. With the Gemma 2 2B TE that Neta Lumina has, such concepts like <code>covered diaper</code> are possible without having any images of that concept, as well as the ability for natural language to be used.</p>        <p>Wetness indicators are now tagged to help stop the wetness indicators from appearing during pantypooping. You can also go and set up the wetness indicators to any color the model recognizes. The most trained colors are yellow and blue, which are common colors used for diaper wetness indicators, but if you use Neta Lumina, you can also set it to any color it knows.</p>        <ul>        <li>Added handmade natural language captions to <code>diaper</code> and <code>peed diaper</code> concepts images in the dataset.</li>            <ul>                <li>The <code>diaper</code> and <code>peed diaper</code> concepts are hand captioned.</li>                <li>Other images and newer images will use a combination of three VLMs summarized into one caption with tag grounding: JoyCaption Beta 1, MiMo Vision 7B, and Gemma 3 27B. This will be done on V17.</li>                <li>This is to ensure that this LoRA is easily usable on models such as ~~SD3.5~~ Neta Lumina and any other future natural language models.</li>            </ul>        <li>Added artist tags to all identifiable artworks due to an accidental discovery where AIDL also learned artist styles whn there was enough data.</li>        <li>Added new diaper design: <code>merries</code><li>        <li>Tagged <code>wetness indicator</code>s as an attempt to separate wetness indicators from soiling (apparent when doing panty pees/poops).</li>        <li>Redone prints and made sure to catch as many prints as possible, including the use of <code>polka dot diaper</code>, <code>orientation indicator</code>, <code>size indicator</code></li>        <li>Added LoRA Manager metadata</li>        </ul>    </div></div>",
+    "trainedWords": [
+      "diaper",
+      "peed diaper",
+      "poopy diaper",
+      "pull-ups",
+      "peed pull-ups",
+      "pooping self",
+      "poopy underwear",
+      "poopy shorts",
+      "poopy pants",
+      "open diaper",
+      "covered diaper",
+      "diaper peek",
+      "@cakies22, oekaki",
+      "@omulettes",
+      "@mizusoba",
+      "@mogurin",
+      "@kyoju",
+      "@dia75",
+      "@bunnymim",
+      "@gogo \\(201177\\)",
+      "@ararii",
+      "@tokiha \\(ruinluin\\)",
+      "@azure_sunset4",
+      "@shibachaya",
+      "@alkali",
+      "@road again",
+      "@hc68o33",
+      "@stolas \\(lemegeton\\)"
+    ],
+    "customImages": []
+  },
+  "tags": [
+    "diaper",
+    "abdl",
+    "omorashi",
+    "scat"
+  ],
+  "modelDescription": "<style>  .main-div {    background-color: #9999ff; padding: 20px; border-radius: 10px; margin-bottom: 20px;  }  h1 {    color: #24292e;    margin-top: 0;    font-size: 25px;  }  h2 {    color: #24292e;    margin-top: 0;    font-size: 22px;  }  h3 {    color: #24292e;    margin-top: 0;    font-size: 20px;  }  h4 {    color: #24292e;    margin-top: 0;    font-size: 15px;  }  p {    color:#24292e;  }  a {    color: #0366d6;    text-decoration: none;  }  .section-div {    background-color: white;    padding: 15px;    border-radius: 8px;    margin: 15px 0;    box-shadow: 0 2px 4px rgba(0,0,0,0.1);  }  .subsection-div {    background-color: #bbb;    padding: 15px;    border-radius: 8px;    margin: 15px 0;    box-shadow: 0 2px 4px rgba(0,0,0,0.1);  }  ul {    margin-bottom: 0;    color: #24292e;  }  ol {    margin-bottom: 0;    color: #24292e;  }  td {    margin-bottom: 0;    color: #24292e;  }  code {    background-color: #333;    padding: 3px 6px;    border-radius: 4px;  }  table {    color: #24292e;  }</style><div class=\"main-div\">  <h1>AIDL: OmoAIShi</h1>  <div style=\"background-color: #ffff99; padding: 20px; border-radius: 10px; margin-bottom: 20px;\">  <img src=\"docs/Images/v16-thumb.png\">  <h3>V16: Tag Restructure Edition<br>For Lumina 2, Illustrious 2, NoobAI VPred</h3>  <div style=\"background-color: #ff9999; padding: 20px; border-radius: 10px; margin-bottom: 20px;\">  <h3>This model is designed for **NSFW** purposes, trained and tested on DRAWN CHARACTERS ONLY.</h3>  <h3>PUBLIC ALPHA. GENERATIONS DO NOT REPRESENT THE FINAL MODEL.</h3>  </div>    <a href=\"https://gitgud.io/mrpopsalot/aidl-omoaishi\"><img alt=\"Static Badge\" src=\"https://img.shields.io/badge/GitLab-orange?logo=gitlab&logoColor=white&style=for-the-badge\"></a>    <a href=\"https://huggingface.co/mrpopsalot/aidl-omoaishi\"><img alt=\"Static Badge\" src=\"https://img.shields.io/badge/HuggingFace-yellow?logo=huggingface&logoColor=white&style=for-the-badge\"></a>    <a href=\"https://allthefallen.moe/forum/index.php?resources/aidl-omoaishi.7465/\"><img alt=\"Static Badge\" src=\"https://img.shields.io/badge/AllTheFallen-pink?logo=outlook&logoColor=white&style=for-the-badge\"></a>    <img alt=\"GitLab License\" src=\"https://img.shields.io/gitlab/license/mrpopsalot%2Faidl-omoaishi?gitlab_url=http%3A%2F%2Fgitgud.io&style=for-the-badge\">  </div>  <div class=\"section-div\">    <h2>Overview</h2>    <p>OmoAIshi is the most comprehensive multi-concept LoRA for diaper wearing and using, able to comprehend multiple concepts previously unfulfilled by base models and other LoRAs. LyCORIS LoCon is used instead of the regular LoRA format as it allows multiple concepts to be trained within the same model. No more girls in panties!</p>    <ul>      <li><b>Trained Models:</b></li>      <ul>        <li><b>NoobAI-XL VPred:</b> Laxhar/noobai-XL-Vpred-1.0 @ 1024</li>        <li><b>Illustrious 2:</b> OnomaAIResearch/Illustrious-XL-v2.0 @ 1536</li>          <ul>          <li>Also works on 1024px, as well on Illustrious 1, 0.1, and NoobAI EPS</li>          </ul>        <li><b>Lumina Image 2:</b> neta-art/neta-lumina 1.0 @ 1024</li>      </ul>      <li><b>Developed by:</b> Mr Pops Alot</li>      <li><b>Model type:</b> LyCORIS, Full Preset (LoRA for Lumina 2)</li>      <li><b>License:</b> Apache 2.0</li>    </ul><br>    <p>This is the first LoRA of the AIDL series. <b>AIDL: The AI Diaper LyCORIS</b> is intended to a series of LoRAs that teaches models about diapers and other ABDL related concepts.</p>  </div>  <div class=\"section-div\">    <h2>Showcase</h2>    <p><img src=\"docs/Images/EasyGrid_00001_.png\"></p>  </div>  <div class=\"section-div\">    <h2>Features</h2>    <ul>      <li><b>Tape-on diapers:</b> Two and four taped, printed and plain diapers are supported. Diapers from specific brands and designs are also trained when noticed. Note that these are not fully trained and may not be fully accurate.</li>      <li><b>Pull-up diapers:</b> Pull-ups are also trained and can be prompted in various designs.</li>      <li><b>Peed diapers:</b> Wet diapers of various stages, from slightly wet to completely leaking. Speaking of leaks, they now correctly go at the side of the thighs.</li>      <li><b>Diaper and panty pooping:</b> Soiling of various stages, including blowouts have also been trained, as with double pee and poo soiling as well. Pants, shorts, and dress pooping is also covered. Compatibility is not guaranteed (there is less training data). You may see some exposed scat.</li>      <li><b>Open diapers:</b> Open diapers are now possible to do, complete with wetness and mess in the pads if prompted.</li>      <li><b>Consistent results across different character positions:</b> This model has been trained on a variety of styles and poses, ensuring correct generation on many poses and whatever artstyle you want to use, including standing, squatting, lying down, from below upskirts, and more</li>    </ul>  </div>  <div class=\"section-div\">    <h2>How to Use</h2>    <div class=\"subsection-div\">      <h3>Diapers</h3>      <p>Use the general term: <code>diaper</code>, or <code>in a diaper</code>.</p>      <ul>      <li>You can also specify the tape count using <code>two taped</code>, or <code>four taped</code>.</li>      <li>Prompting <code>plain diaper</code> will get you a diaper with no designs, while <code>print diaper</code> will get you a printed diaper.</li>      <li><code>front-print diaper</code> and <code>back-print diaper</code> can also be used to get you a print on the front or back. Use <a href=\"https://danbooru.donmai.us/tags?commit=Search&search%5Bhide_empty%5D=yes&search%5Bname_or_alias_matches%5D=%2A_print&search%5Border%5D=count\">Danbooru's list of prints</a>.</li>      <li>Add in wetness indicators by prompting <code>yellow wetness indicator</code> or <code>blue wetness indicator</code>.</li>      </ul>      <p>Specific diaper styles:</p>      <ul>        <li><code>pampers</code></li>        <li><code>huggies</code></li>        <li><code>luvs</code></li>        <li><code>abuniverse</code></li>        <li><code>merries</code></li>      </ul>    </div>    <div class=\"subsection-div\">      <h3>Pull-Ups</h3>      <p>Use <code>pull-ups</code>. Same prompting style as with diapers, but replace <code>diaper</code> with <code>pull-ups</code>.</p>    </div>    <div class=\"subsection-div\">      <h3>Diaper Soiling</h3>      <p><b>Peed-in diapers:</b> Use the general <code>peed diaper</code> tag to get wet diapers. From there, you can also go and set how soiled the diaper should be. Without it, you might get unexpected diaper blowouts.</p>      <ol>        <li><code>wet diaper</code></li>        <li><code>soaked diaper</code></li>        <li><code>drenched diaper</code></li>        <li><code>leaking diaper</code>, <code>pee on legs</code></li>      </ol>      <br>      <p><b>Poopy diapers:</b> Use the general <code>poopy diaper</code> tag to get messy diapers. From there, you can also go and set how soiled the diaper should be. Without it, you might get unexpected diaper blowouts.</p>      <ol>        <li><code>messy diaper</code></li>        <li><code>packed diaper</code></li>        <li><code>loaded diaper</code></li>        <li><code>diaper blowout</code>, <code>poop on legs</code></li>      </ol>      <br>      <p><code>smell</code> can also be added to your prompt to give off that soiled smell effect.</p>      <p>If you want more soiled diapers, you can increase the weight of the tag to suit your needs. Tags works well between 0.9-1.5 weight.</p>      <img src=\"docs/Images/unidesSoil_00002_.png\">      <img src=\"docs/Images/unidesSoil_00004_.png\">    </div>    <div class=\"subsection-div\">      <h3>Pantypooping</h3>      <p>Use the term <code>pooping self</code>. You can also use the terms <code>poopy underwear</code> or <code>poopy clothes</code> to have either their underwear be soiled or their lower clothes (like their dress or pants) to be soiled.      </p>      <img src=\"docs/Images/unidesSoil_00005_.png\">      <p>Multiple underwear types supported!</p>      <p>For diapers please use the poopy diaper tag. Pooping self can also work but poopy diaper is the recommended tag for diaper pooping.</p>    </div>    <div class=\"subsection-div\">      <h3>Open Diapers</h3>      <p>Use the term <code>open diaper</code> to get open diapers.</p>      <ul>        <li>Add in <code>peed diaper</code> and <code>poopy diaper</code> respectively if you want the diaper to be wet and messy.</li>        <li>As well, add your respective genitals such as <code>pussy</code> or <code>penis</code> and the uncensored tag</li>      </ul>      <img src=\"docs/Images/openDiaper_00000_.png\">    </div>  </div>  <div class=\"section-div\">    <h2>Bias, Risks, and Limitations</h2>    <p>The model currently struggles with the following concepts:</p>    <ul>      <li>Non-printed and colored pull-ups</li>      <li>Open diapers</li>      <ul>        <li>Open diapers do not have as much data as other concepts, and will struggle to have proper diaper structure.</li>      </ul>      <li>Unworn diapers (both clean and used)</li>      <li>Clothespooping</li>      <li>Technical positions such as falling</li>    </ul>    <br>    <p>As well, there may be some limitations with this LoRA:</p>    <ul>      <li>Panty poops may happen unprompted if <code>smell</code> is prompted</li>      <li>Peed diapers may have poopy diaper bulges unless poopy diaper is put in the negatives</li>      <li>There is currently no training data for poopy pull-ups, and is using the fallback pantypooping data.</li>      <li>Poopy diapers may have unprompted blowouts.</li>      <li>In the Lumina 2 LoRA, you need to prompt for \"wet diaper\" as well with \"peed diaper\" to get a pee soaked pad.</li>      <li>Although you can use Rouwei Gemma in the SDXL variants, you'll only get pull-ups</li>    </ul>  <p>Increase the strength of the LoRA if needed to achieve your desired results.</p>  <div style=\"background-color: #ff9999; padding: 20px; border-radius: 10px; margin-bottom: 20px;\">    <p> I am in no way responsible for any content that the user creates with malicious intent while using any of these models. The user assumes complete responsibility for the misuse of these tools and any unethical content created with them. <b>Follow all local, state, and federal laws as well as online rules regarding this content and generations with this LoRA. I am not responsible if you are arrested or banned from an online service for using this.</b></p>    <p><b>This LoRA violates CivitAI's terms of service (their TOS does not allow diapers nor scat). Do not upload generations from AIDL onto CivitAI.</b></p>  </div>  </div>  <div class=\"section-div\">    <h2>Training Details</h2>    <p>Over 1000 images have been used in the dataset. Images used are <b>exclusively drawn 2D</b> art.</p>    <p><b>No synthetic data has been used</b> for the model. Tagging (Danbooru tag style) is <b>all manually done</b>, as automatic taggers do not recognize the concept in question well. Image captioning (NL support) uses a list of manual captioned images and (planned for V17) JoyCaption Beta 1 and Gemma 3 27B outputs with manual Tag grounding.</p>    <div class=\"subsection-div\">      <h3>Training Procedure</h3>      <p>Training configs are provided with the weights, but as a TL;DR:</p>      <h4>SDXL-based (Illustrious, NoobAI)</h4>      <ul>        <li>Program: 67372a's Kohya Easy Scripts</li>        <li>Optimizer: CAME</li>        <li>Scheduler: Rex</li>        <li>UNet LR: 3e-4</li>        <li>LoRA type: LoCon</li>        <li>Steps: 36600</li>        <li>Network Rank (Dim/Alpha): 64/1</li>        <li>Precision: Full BF16</li>        <li>Hardware: 1x RTX 3090 24GB</li>      </ul>      <h4>Lumina 2</h4>      <ul>        <li>Program: Kohya sd-scripts SD3 branch</li>        <li>Optimizer: CAME</li>        <li>Scheduler: Rex</li>        <li>DiT LR: 1e-4</li>        <li>TE LR: None. TE outputs are cached</li>        <li>LoRA type: LoRA Lumina</li>        <ul>          <li>No LoCon needed as they'res no Conv layers.</li>        </ul>        <li>Steps: 36600 (batch 4)</li>        <li>Network Rank (Dim/Alpha): 64/32</li>        <li>Precision: Full BF16</li>        <li>Hardware: 1x RTX 3090 24GB</li>      </ul>    </div>  </div>  <div class=\"section-div\">    <h2>FAQ</h2>    <ul>      <li><b>Will this model be ported to X?</b> This model is planned to have versions trained on NoobAI-XL VPred, Illustrious 2.0 (ILXL2), and Neta Lumina (Lumina 2). More models will come later.      <ul>        <li>Lumina 2 version only works on Neta Lumina AES paths, such as a5nA and 1.0</li>        <li>The Illustrious 2 version works on Illustrious 1, 0.1, and NoobAI-XL Epsilon Pred.</li>        <li>V17 will have a Wan 2.2 5B</li>        <li>No Pony version is planned. Move to NoobXL or ILXL.</li>      </ul></li>      <li><b>Will there be a version for CivitAI?</b> No. I do not trust CivitAI anymore after the release of Illustrious 1.0 and them allowing closed-weight models, not to mention the overbearing censorship against messy diaper content, bad service, and overall greed.      <ul>        <li>V15.1 Update: Definitely not, especially since urine and diaper content have been expressly banned. Oh, and half the time the site crashes when you try to do anything. Good riddance.</li>      </ul></li>      <li><b>Anything else planned?</b>        <ul>          <li>Diaper changes</li>          <li>Open diapers and pull-ups</li>          <li>Diaper covers</li>          <li>Diaper pulls (like panty pulls)</li>          <li>Clean and used unworn diapers</li>          <li>Cloth diapers</li>          <li>Diaper pulls</li>          <li>The rest of the data in my dataset processed and tagged.</li>          <li>Etc.</li>        </ul>      </li>      <ul>    </ul>  </div>  <div class=\"section-div\">    <h2>License</h2>    <p>This LoRA model is provided under the <a href=\"https://huggingface.co/datasets/choosealicense/licenses/blob/main/markdown/apache-2.0.md\">Apache License 2.0</a> license.</p>    <!-- <p>Training configurations and dataset will also be released later. You can use these to further train AIDL.</p> -->  </div>  <div class=\"section-div\">    <h2>Special Thanks</h2>    <ul>      <li>DangoKnight for training some of the test versions of this LoRA</li>      <li>Yatogami for contributing to the dataset.</li>      <li>Hugs288, Lopi99, and others in the NoobAI discord to for helping me get my LoRA training settings right.</li>    </ul>  </div>  <div class=\"section-div\">    <h2>Model Card Contact</h2>    <ul>      <p>I can be contacted on:</p>      <li><b>Matrix:</b> @mr_pops_alot:cutefunny.art</li>      <li><b>GitLab:</b> gitgud.io/mrpopsalot</li>      <li><b>All The Fallen:</b> @MrDoesNothingAlot</li>      <li><b>Discord:</b> 474149843724664854</li>    </ul>  </div></div>",
+  "civitai_deleted": false,
+  "favorite": true,
+  "exclude": false,
+  "db_checked": false,
+  "metadata_source": null,
+  "last_checked_at": 0,
+  "usage_tips": "{\"strength\":1}"
+}

Changelog.MD

@@ -77,13 +77,13 @@
     <!-- <div class="section-div">
         <h2>V18: 2K Edition</h2>
     </div> -->
-    <div class="section-div">
+    <!-- <div class="section-div">
         <h2>V17.1: Qwen3.5 gud Edition</h2>
         <p>Qwen3.5 is GOOD.</p>
         <ul>
             <li>Added natural language captions to all concepts for Anima.</li>
         </ul>
-    </div>
+    </div> -->
     <div class="section-div">
         <h2>V17: Pee-ka-poo Edition</h2>
         <p>New update for AIDL: OmoAIShi (the diaper LoRA). Moar data means more better, right? Here's V17, with 725 more images in the dataset with perfectly balanced concepts.
@@ -99,6 +99,7 @@
                 </ul> -->
             <li>Versions added:</li>
               <ul>
+              <li><b>NEW 2026.04.03</b> Mugen @ 1024</li>
               <li>Anima @ 512</li>
               <li>Chenkin RF @ 1024</li>
               <li>NoobAI RF EQ-VAE @ 1024</li>

Changelog.ja.md

@@ -92,6 +92,7 @@
                 </ul> -->
             <li>新しいバーション：</li>
               <ul>
+              <li><b>NEW 2026.04.03</b> Mugen @ 1024</li>
               <li>Anima @ 512</li>
               <li>Chenkin RF @ 1024</li>
               <li>NoobAI RF EQ-VAE @ 1024</li>

README.ja.md

@@ -16,6 +16,7 @@ base_model:
 - Laxhar/noobai-XL-Vpred-1.0
 - OnomaAIResearch/Illustrious-XL-v2.0
 - neta-art/neta-lumina
+- CabalResearch/Mugen
 ---
 <style>
   .main-div {
@@ -87,7 +88,7 @@ base_model:
   <h1>AIDL: OmoAIShi</h1>
   <div style="background-color: #ffff99; padding: 20px; border-radius: 10px; margin-bottom: 20px;">
   <img src="docs/Images/v017-thumb-JA.png">
-  <h3>V17: Pee-ka-poo Edition<br>Anima, Chenkin RF, NoobAI RF EQ-VAE, NoobAI RF FLUX.2 Vae, Illustrious 2向け</h3>
+  <h3>V17: Pee-ka-poo Edition<br>Anima, Mugen, Chenkin RF, NoobAI RF EQ-VAE, NoobAI RF FLUX.2 Vae, Illustrious 2向け</h3>
   <div style="background-color: #ff9999; padding: 20px; border-radius: 10px; margin-bottom: 20px;">
   <h3>このモデルは**NSFW**画像向け、＊二次元のキャラ＊しか訓練していません.</h3>
   <h3>公開のアルファ版です. 生成された画像は最終的なモデルを表すものではありません.</h3>
@@ -107,6 +108,7 @@ base_model:
       <li><b>訓練されたモデル:</b></li>
       <ul>
         <li><b>Anima:</b> circlestone-labs/Anima @ 512</li>
+        <li><b>Mugen:</b> ChenkinRF/ChenkinNoob-XL-v0.2-Rectified-Flow @ 1024</li>
         <li><b>Chenkin RF:</b> ChenkinRF/ChenkinNoob-XL-v0.2-Rectified-Flow @ 1024</li>
         <li><b>NoobAI-XL RF EQ-VAE:</b> CabalResearch/NoobAI-RectifiedFlow-Experimental/NoobAI-RF-v0.2-Base @ 1024</li>
           <ul>
@@ -348,6 +350,20 @@ base_model:
         <li>精度：Full BF16</li>
         <li>ハードウェア：1x RTX 3090 24GB</li>
       </ul>
+      <h4>Mugen</h4>
+      <ul>
+        <li>ソフト：Kohya sd-scripts Bluvoll fork</li>
+        <li>オプティマイザ：CAME</li>
+        <li>スケジューラ：Constant</li>
+        <li>UNetの学習率：1e-5</li>
+        <li>TE1の学習率：7e-5</li>
+        <li>TE2の学習率：3.5e-5</li>
+        <li>LoRAのタイプ：LoKr フル行列</li>
+        <li>エポックの数：16</li>
+        <li>ネットワークのランク (次元/アルファ/因数分解): 512/1/4</li>
+        <li>精度：Full BF16</li>
+        <li>ハードウェア：1x RTX 3090 24GB</li>
+      </ul>
       <!-- No lumina version trained -->
       <!-- <h4>Lumina 2</h4>
       <ul>

README.md

@@ -16,6 +16,7 @@ base_model:
 - Laxhar/noobai-XL-Vpred-1.0
 - OnomaAIResearch/Illustrious-XL-v2.0
 - neta-art/neta-lumina
+- CabalResearch/Mugen
 ---
 <style>
   .main-div {
@@ -87,7 +88,7 @@ base_model:
   <h1>AIDL: OmoAIShi</h1>
   <div style="background-color: #ffff99; padding: 20px; border-radius: 10px; margin-bottom: 20px;">
   <img src="docs/Images/v017-thumb-EN.png">
-  <h3>V17: Pee-ka-poo Edition<br>For Anima, Chenkin RF, NoobAI RF EQ-VAE, NoobAI RF FLUX.2 Vae, Illustrious 2</h3>
+  <h3>V17: Pee-ka-poo Edition<br>For Anima, Mugen, Chenkin RF, NoobAI RF EQ-VAE, NoobAI RF FLUX.2 Vae, Illustrious 2</h3>
   <div style="background-color: #ff9999; padding: 20px; border-radius: 10px; margin-bottom: 20px;">
   <h3>This model is designed for **NSFW** purposes, trained and tested on DRAWN CHARACTERS ONLY.</h3>
   <h3>PUBLIC ALPHA. GENERATIONS DO NOT REPRESENT THE FINAL MODEL.</h3>
@@ -110,6 +111,8 @@ base_model:
           <ul>
           <li>Note that the base model is also trained on 512px.</li>
           </ul>
+        <li><b>Anima:</b> circlestone-labs/Anima @ 512</li>
+        <li><b>Mugen:</b> ChenkinRF/ChenkinNoob-XL-v0.2-Rectified-Flow @ 1024</li>
         <li><b>Chenkin RF:</b> ChenkinRF/ChenkinNoob-XL-v0.2-Rectified-Flow @ 1024</li>
         <li><b>NoobAI-XL RF EQ-VAE:</b> CabalResearch/NoobAI-RectifiedFlow-Experimental/NoobAI-RF-v0.2-Base @ 1024</li>
           <ul>
@@ -354,6 +357,20 @@ base_model:
         <li>Precision: Full BF16</li>
         <li>Hardware: 1x RTX 3090 24GB</li>
       </ul>
+      <h4>Mugen</h4>
+      <ul>
+        <li>Program: Kohya sd-scripts Bluvoll fork</li>
+        <li>Optimizer: CAME</li>
+        <li>Scheduler: Constant</li>
+        <li>UNet LR: 1e-5</li>
+        <li>TE1 LR: 7e-5</li>
+        <li>TE2 LR: 3.5e-5</li>
+        <li>LoRA type: LoKr Full Matrix</li>
+        <li>Epochs: 16</li>
+        <li>Network Rank (Dim/Alpha/Factor): 512/1/4</li>
+        <li>Precision: Full BF16</li>
+        <li>Hardware: 1x RTX 3090 24GB</li>
+      </ul>
       <!-- No lumina version trained -->
       <!-- <h4>Lumina 2</h4>
       <ul>
@@ -457,4 +474,4 @@ base_model:
   <p>I do not have a SubscribeStar, as they do not seem to want to approve my account. I have never and will never open up a Patreon, as they are infamous for their ban waves and attacks against the NSFW community. Please support me on these alternative services.</p>
   </div>
   </div>
-</div>
+</div>