原作者：上善无形、ZipZipPipe。 协议：<a target="_blank" rel="ugc" href="https://creativecommons.org/licenses/by-nc-sa/4.0/deed.zh-hans">CC BY-NC-SA 4.0</a> 严禁用于商业用途先叠甲免责声明 ：下载并使用本LoRA自动表示已认真阅读免责声明并同意协议内容本模型为粉丝同人创作，基于 AI绘图/LoRA 技术制作，并非深度求索（DeepSeek）公司官方出品，亦不代表其立场、观点或授权。“DeepSeek”及相关名称、标识、Logo、图形等权利均归其权利人所有。本模型仅用于个人学习、技术研究、同人交流，不构成商标性使用，亦不用于任何商业用途。本模型采用 CC BY-NC-SA 4.0 许可发布。该许可仅涉及版权意义上的使用、复制与改编，不授予任何商标权利。“DeepSeek”相关商标权利仍归原权利人所有。使用者不得将本模型用于： 商业盈利、广告宣传、付费下载或变相销售；冒充 DeepSeek 官方内容或暗示官方合作；生成损害品牌形象的或其他侵权内容。因使用者违反上述约定或自行生成、传播相关内容所产生的法律责任，由使用者自行承担，与作者无关。若相关权利人认为本模型存在侵权或不当使用，请及时联系，本人将在收到通知后及时下架、删除或修改相关内容。本模型按“现状”提供，不提供任何明示或暗示担保。作者不对因使用本模型产生的任何直接或间接损失负责。 简介：该lora是凭收集到的少量女仆装鲸鱼娘图片，以anima-base-1.0为底模进行训练，用于让鲸鱼娘更符合现下流行女仆特点的简易lora。 触发词：@deepseek此版本对tags变化敏感，建议使用简单tags，较多与女仆装鲸鱼娘形象重合的外观服装描述会减弱效果，提高权重可以有效解决。该lora为出于测试目的的初版lora，训练数据集较少，未进行大量测试，可能会因为训练不足的原因而效果较差。效果不明显考虑提高权重至1.5<s>训练偶遇围裙鲸鱼图案，🦌尽全力无法战胜。</s> English Version: Original authors: Shangshan Wuxing, ZipZipPipe License: CC BY-NC-SA 4.0Disclaimer:By downloading and using this LoRA, you automatically confirm that you have carefully read this disclaimer and agree to its terms.This model is a fan-made creation based on AI image generation / LoRA technology. It is not an official DeepSeek product, nor does it represent the company’s stance, views, or endorsement. “DeepSeek” and related names, marks, logos, and graphics are the property of their respective owners. This model is intended solely for personal study, technical research, and fan exchange. It does not constitute trademark use and is not intended for any commercial purpose. This model is released under the CC BY-NC-SA 4.0 license. This license only covers copyright-related use, reproduction, and adaptation, and does not grant any trademark rights. All trademark rights related to “DeepSeek” remain with their original owners.Users may not use this model for: commercial profit, advertising, paid downloads, or disguised sales; impersonating official DeepSeek content or implying official cooperation; or generating content that damages the brand image or infringes upon others’ rights. Any legal liability arising from users violating the above terms or from content they generate or distribute shall be borne by the users themselves and is unrelated to the author.If any rights holder believes this model constitutes infringement or improper use, please contact me promptly. I will remove, delete, or modify the relevant content upon notification. This model is provided “as is,” without any express or implied warranty. The author is not liable for any direct or indirect losses resulting from use of this model.Description:This LoRA was trained on a small set of collected images of a maid-outfit whale girl using anima-base-1.0 as the base model. It is a simple LoRA intended to make the whale girl better match the current popular maid style.Trigger word:@deepseekThis version is sensitive to tag changes. It is recommended to use simple tags. Including too many appearance and clothing tags that overlap with the maid-outfit whale girl image will weaken the effect. Increasing the weight can effectively solve this.This LoRA is an initial test version with a small training dataset and has not been extensively tested. It may perform poorly due to insufficient training.If the effect is not obvious, consider increasing the weight to 1.5.

Files with SHA256: 85fe6d590e8a38c6abafc54ec0540b253eff524c29553bd1c4547efeb8a52fba