<h3 id="character-table-rqhy7ihz9">CHARACTER TABLE</h3><ul><li></li></ul><ul><li>Sovetskaya Rossiya (苏维埃·罗西亚)</li><li>Ying Swei (应瑞)</li><li>Zara (扎拉)</li></ul><h3 id="faq-wmm6fcqav">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<h3 id="-2p109z30z"></h3>

Azur Lane

Arkhangelsk.safetensors

Arkhangelsk azur lane

Arkhangelsk

Ognevoy

Rossiya

Zara

Swei

ARKHANGELSK \(AZUR LANE\), ARKHANGELSK REGULAR WEAR, ARKHANGELSK GRANDIOSE GETAWAY, ARKHANGELSK SUPPORTING SORCERESS

ARKHANGELSK \(AZUR LANE\),  ARKHANGELSK REGULAR WEAR, ARKHANGELSK GRANDIOSE GETAWAY, ARKHANGELSK SUPPORTING SORCERESS

Files with SHA256: 190220165768581c69e2db9e78f7967f01b6aa540a0cdc4ecdf0b588e584bdd7