本页面包含 MOSS 和 Anima 两个版本，请选择对应的使用说明。 This page includes both MOSS and Anima versions. Please follow the instructions for your model._______________________________________________________________________________________<h2 id="recommended-settings">推荐参数/Recommended Settings</h2>FOR ANIMA:<ul><li>触发词/Trigger word: @ttaistyle</li><li>权重/Recommended weight:0.7~1</li><li>Sampler: er_sde + beta OR euler + normal</li><li>CFG:4</li><li>STEPS:30</li></ul><h3 id="for-moss">FOR MOSS：</h3>复杂场景，请提高CFG和步数，SFT例外。For complex scenes, increase the CFG and sampling steps,except when using SFT.<ul><li>触发词/Trigger word: @ttaistyle</li><li>权重/Recommended weight:1~1.2</li><li>Sampler: er_sde + beta OR euler + simple</li><li>CFG:3-5</li><li>STEPS:30-50</li></ul><h3></h3>____________________________________________________________________________________<h2 id="moss">MOSS</h2><h3 id="v1.1">V1.1</h3><ul><li>和ANIMA同数据集训练，感觉依然欠拟合，我大概会再作尝试。</li><li>Trained on the same dataset as ANIMA. It still feels underfit, though, so I’ll probably keep experimenting.</li><li>请测试，如果有什么问题，请留言告知。Please test it and let me know in the comments if you encounter any issues.</li></ul>____________________________________________________________________________________<h2 id="anima">ANIMA</h2><h3 id="v2">V2：</h3><ul><li>新增了些许数据集。</li><li>修改了配方和使用了新的训练器，整体表现更好，指令遵从度改善，光影、画面、阴影也有一定改善。</li><li>解决了马赛克容易出现的问题，但由于数据集基本全是有马赛克，所以你仍然要透过提示词稍作控制。</li><li>added a small amount of new training data.</li><li>Updated the training settings and switched to a new trainer, resulting in better overall performance, improved prompt adherence, and noticeable improvements in lighting, image quality, and shadows.</li><li>Fixed the issue of mosaics appearing too frequently.However, since most of the training dataset contains mosaic censorship, you'll still need to use prompts to keep it under control.</li></ul>________________________________________________________________________

Files with SHA256: 6b2a1b54ae8ec4947623b2dd902c13b3c95e0a6f6d841cfc7debc3be21bc15f9