V2 Update:

In this V2 update, compared with V1, the number of training set samples has been reduced and the samples have been re-filtered to ensure style consistency and to match the artist's recent art style.

This V2 update includes two model variants:

• V2 Less Step: trained with fewer steps.

• V2 More Step: trained with more steps.

It is recommended to use er sde/beta57

The difference between them: More Step is closer to the original artist's style; Less Step broadly approximates the original artist's style while offering a different feel in the details compared with More Step.

Everyone is welcome to try whichever you prefer. It would be even better if you could share your generated images. Thanks for your continued support and feedback
A little complaint: This V2 update was scrapped and redone 4 times from start to finish — constantly adjusting training parameters, modifying the training set, and testing results. It took a whole week, and I wasn't satisfied until this version. My electricity bill is burning up 😭

本次 V2 更新相比 V1 减少了训练集样本数量，并进行了重新筛选，以保证风格统一且符合近期画风。

本次 V2 更新包含两个模型版本：

• V2 Less Step：训练步数更少。

• V2 More Step：训练步数更多。

推荐使用 er sde/beta57

两者的区别：More Step 更接近原画师画风；Less Step 在大体上接近原画师画风的同时，在细节方面会与 More Step 有不一样的感觉。

欢迎大家根据喜好选择试用，要是能返图就更好了，感谢大家一直以来的支持与反馈。
最后吐槽一下：这次 V2 更新前前后后推倒重来了 4 次，不断调整训练参数、修改训练集、测试结果，花了 1 周时间，一直没有得到满意的结果，直到这次才终于满意。电费急剧燃烧中 😭

=============================================================

This model is trained on images created by the artist Greyash, aiming to replicate her unique artistic style.

If the original artist has any objection to this model, please contact me via private message and I will remove it immediately and unconditionally.

## ⚠️ Important Notice

- This model is intended for learning and research purposes only. All rights to the artistic style belong to Greyash.

- Commercial use is strictly prohibited (including but not limited to selling generated images, or using on commercial generation services).

- Unauthorized redistribution or uploading to other platforms is not allowed.

- Users are solely responsible for any consequences arising from the use of this model.

## 💡 Usage Guide

- Trigger word: @Greyash

- Recommended weight: 1.0

- Recommended base model: Anima

Finally, I would like to express my sincere gratitude to Greyash for her wonderful artwork.
----------------------------------------------------------------------------------------------------------
本模型是基于艺术家 Greyash 的图像进行训练的，风格旨在还原其个人特色。

若原画师本人对本模型的存在有任何异议，请通过 私信 联系我，我将随时无条件删除下架。

## ⚠️ 重要声明

- 本模型仅用于**学习与研究目的**，该画风的所有权利归 Greyash 所有。

- 严禁将本模型用于**任何商业用途**（包括但不限于销售生成的图像、使用于商业生成服务平台）。

- 未经许可，禁止将本模型文件转载或上传至其他平台。

- 用户使用本模型生成内容所产生的任何后果，由用户自行承担。

## 💡 使用指南

- 触发词 @Greyash

- 推荐权重1.0

- 推荐底模Anima

最后，由衷感谢 Greyash 创作的精彩作品！
------------------------------------------------------------------------------------------------------
这位艺术家的风格偏向于高曝光，所以有时候对画面光照会有影响，也许可以有增强光照的效果？（bushi）感谢每一位尝试使用本模型的朋友！如有任何问题或建议，欢迎留言或私信我。希望你们玩得开心！
This artist's style leans toward high exposure, so it can sometimes affect the lighting in the images—maybe it could even enhance the lighting effect? (just kidding)

Thanks to everyone who gives this model a try! If you have any questions or suggestions, feel free to leave a comment or DM me. Hope you all have fun!