Please, check the description, there is a little bit of useful information.<h3>Indroduction:</h3>This lora is supposed to represent hyouuma's artstyle.Model used for previews: <a target="_blank" rel="ugc" href="https://civarchive.com/models/4855/anythingelse-v4">Anything v4</a> + <a target="_blank" rel="ugc" href="https://civarchive.com/models/9409/or-anything-v5">Anything v5 </a>(v1)Model used for previews: <a target="_blank" rel="ugc" href="https://civarchive.com/models/6174/divineelegancemix">DivineElegance v5 </a>(v2)Embeddings used for preview (v1): <a target="_blank" rel="ugc" href="https://civarchive.com/models/7808/easynegative">EasyNegative </a>and <a target="_blank" rel="ugc" href="https://huggingface.co/Cordeliya/animefull-latest/resolve/main/bad-hands-3.pt">Bad-Hands-3</a>Be aware! Don't use EasyNegative with Style LoRAs, it may ruin the style!Works pretty good with side character LoRAs<h3>Artist's main features:</h3><ul><li>Very soft and silkish skin</li><li>Beautiful asses, pussies and anuses</li><li>Pantyhose and thighhighs</li><li>Full-anime faces</li></ul><h3>Recommendations:</h3><ul><li><code>hyouuma</code> activation tag (v1)</li><li>No activation tag (v2)</li><li>1.0 weight</li></ul><h3>Credits:</h3>Trained on NAI Anime Full. Dataset of 250 pictures. 18 Epochs. (v1) Clip Skip - 2Dataset of 250 pictures. 20 Epochs. (v2) Clip Skip - 2Really big thanks to:<ul><li><a target="_blank" rel="ugc" href="https://www.pixiv.net/en/users/20235288">hyouuma (H, 20235288) </a>for beautiful artwork</li><li>Some good folks from Discord server</li><li>You, for reading, downloading and reviewing this LoRA</li></ul>

hyouuma - Style LoRA

Version changes:<ul><li>Less weight</li><li>Better style representation</li><li>New concepts (still need improvements)</li><li>Trigger word needed no more</li></ul>

Files with SHA256: 518e13becdd36136924f3750239b8b4d788b31a34a104bb8139244336071995e