<h1 id="hatawagashi_onionmix_style">hatawagashi_onionmix_style</h1>Trigger word： @hatawgs_onionmix<h2>中文介绍</h2>模型名称：hatawagashi_onionmix_style 触发词：@hatawgs_onionmix 模型简介：二次元混合画风 LoRA，融合多位画师风格特征。 画风主体为 hata4564 和 wagashi 两位画师风格混合，搭配其余多位画师风格特征调和成型。本 LoRA 炼制所用数据集，由光辉系列模型生成图片制作。推荐生成参数：迭代步数 30、CFG 3.5‑5、采样器 euler_a。免责声明：本模型仅用于个人 AI 绘画创作学习使用，请勿用于商用、侵权及违规场景。模型画风由多画师风格混合训练，仅供艺术创作参考，若产生版权纠纷，使用者自行承担相关责任。若对应画风所有者提出相关意见，本人将立刻下架、删除本 LoRA 模型。<ul><li>版本更新 V2 画风更加符合预期；我分别在 Anima‑b1 和 Anima‑2.9B 基底模型上尝试使用本 LoRA，都可以正常工作，出图效果都很不错。</li><li>以上介绍为 AI 生成</li></ul><hr /><h2 id="english-introduction">English Introduction</h2>Model Name: hatawagashi_onionmix_style Trigger word: @hatawgs_onionmixModel Description: An anime‑style mixed LoRA that incorporates stylistic features from multiple artists. Its core aesthetic is a blend of art styles from hata4564 and wagashi, further balanced with stylistic elements from several other creators. The dataset used for training this LoRA consists of images generated by the Radiance series models.Recommended generation parameters: Steps 30 CFG scale 3.5‑5, Sampler euler_a.Disclaimer: This model is for personal AI‑art study purposes only. Commercial use, copyright‑infringing and any other improper usage are prohibited. Trained with mixed art styles of various artists, this LoRA serves purely as a reference for artistic creation. Users shall bear full responsibility for any potential copyright disputes. Should the owners of the original art styles raise objections, I will immediately take down and delete this LoRA model.<ul><li>Version Update V2 The art style now better matches expectations. I have tested this LoRA on the Anima‑b1 and Anima‑2.9B base models. It works properly on both base models and yields fairly good generation results.</li></ul><ul><li>The above description is AI‑generated</li></ul>

hatawagashi_onionmix style

版本更新 V2画风更加符合预期；我分别在 Anima‑b1 和 Anima‑2.9B 基底模型上尝试使用本 LoRA，都可以正常工作，出图效果都很不错。Version Update V2 The art style now better matches expectations. I have tested this LoRA on the Anima‑b1 and Anima‑2.9B base models. It works properly on both base models and yields fairly good generation results.

Files with SHA256: 0a8e73ec11fe810d2e14d51357137f30ee833af554e233121cf8ad92aaaae608