Model info

Fine-tuned on Anima-Preview.

This LoRA is fine-tuned on a dataset of 300+ images. It can recreate The Dahlia using the trigger word Constance alone.

Param Settings

• Recommended weight: 0.8 - 1.0

• example prompt :

masterpiece, best quality, score_7, safe, anime, Constance, 1girl, solo, hat, blush, dress, looking at viewer, cowboy shot, detailed light. She is smiling confidently at the viewer. She is dancing ballet under spotlight. The background is in the theater.

Tips

• Official design is bound to the following tags: hat, dress, gloves, high heels.

• If you do not want the tail, add tail to the negative prompt.

• Due to the fact that she has large breasts, and this is further amplified by Anima, the size may become... a bit too much. To balance the visual effect, you may consider adding large breasts to the negative prompt, or adding medium breasts to the positive prompt.

Others

In addition to LoRA based on Anima , I have also trained versions based on IllustriousXL, NoobAI, AnimagineXL, PonyXL, WAI-IllustriousXL, and ZukiAnimeIll.

As this is a fresh release, I am currently experimenting with new parameter ratios to find the optimal balance.

Thank you for your support!

模型信息

该模型在Anima-Preview基础上，300+ 张图片的数据集微调得到。此LoRA使用触发词Constance 即可还原大丽花。

参数设置

• 推荐权重： 0.8 - 1.0

• 示例prompt:

masterpiece, best quality, score_7, safe, anime, Constance, 1girl, solo, hat, blush, dress, looking at viewer, cowboy shot, detailed light. She is smiling confidently at the viewer. She is dancing ballet under spotlight. The background is in the theater.

Tips

• 官方服设绑定在以下标签：hat, dress, gloves, high heels。

• 如果你不想要尾巴，在负面提示词加 tail。

• 由于大丽花本来胸大，加之上Anima 的影响可能会有点…太大了。为了平衡视觉效果，可以考虑在负面提示词中加入 large breasts，或在正面提示词中加入 **medium breasts** 。

小记

除了基于Anima的版本，我还训练了基于IllustriousXL，NoobAI, AnimagineXL，PonyXL，WAI-IllustriousXL，ZukiAnimeIll的版本。

如果有任何反馈意见欢迎评论，以便我做出调整，谢谢。我会定期在社交媒体发布图片，可在对应平台取原图。

Bilibili账号：https://space.bilibili.com/12595237?spm_id_from=333.1007.0.0