<h3 id="character-table-29rdvqne4">CHARACTER TABLE</h3><ul><li>알벨다 (Albelda)</li><li>쿤 아게로 아그니스 (Khun Aguero Agnis)</li></ul><h3 id="faq-gwo0171eo">FAQ</h3><ul><li>Models are too big!</li></ul>Yes, so it is recommended to use an online platformThe 64-channel pony model is very large, but it can better support multiple skins<ul><li>Does the character only need one label?</li></ul>Yes, it is recommended to use a single role tag. Multiple tags means the semantics are not properly separated and absorbed.<ul><li>Can the character change clothes?</li></ul>Yes, the character can change clothesGenerally there is a label of XX Regular Clothes to absorb regular clothesIf the character has skin, there will be additional labels to absorb.<ul><li>Does the character support NSFW?</li></ul>Yes, the nude data accounts for about 25%-40% of the dataset,<ul><li>What is the best parameter setting?</li></ul>There is no best parameter, nor recommended parameter.In fact, each basic model has its own G-spot, which needs to be explored slowly.<ul><li>The character does not look like the character in the game</li></ul>It is possible, because the training mainly uses the masterpieces on pixiv.<ul><li>The character label is completely invalid</li></ul>Okay, I may have uploaded the wrong model, thank you for pointing it out, I will replace the correct model.<ul><li>When to train missing characters</li></ul>Until I collect more than 20 high-quality works,Computing is not a problem for me, but good data is hard to collect

Tower of God

Ha Yura.safetensors

Ha Yura

Files with SHA256: f589222d15a576a3011d84d2ca05584d62ae33d0db42ad5e58385afec7d80be0