这个模型使用了超过100张高质量的原始图片进行训练。
这个模型拥有很好的画风模仿效果。TV版 《草莓棉花糖》 图片素材没有混入其中，得到纯净的Sweet Rose (Barasui)风格LORA。
Animagine XL V3版本：https://civarchive.com/models/105656
illustriousXL 版本：https://civarchive.com/models/889207
碎碎念：
Pony Diffusion V6 XL是一个优秀的底模，不容否认确实有概率出不错的图片。
但是对于动漫风格，Pony最大的缺点就是混杂了太多3D和2.5D素材，这会导致纯2D风格的lora并不能稳定的输出指定风格，如果是有同名角色“撞车”的情况下就更明显了。
请参看：安娜柯波拉(Ana Coppola) 的例图。
碎碎念2：（仅中文）
一定有人说：这肯定是UP你的技术不行，你的提示词咒语不行，训练方式方法不行……
我想说的是：你说的对！
推荐权重：0.6~1.0
推荐模型：Pony Diffusion V6 XL
--- --- --- --- --- --- --- ---
This model used more than 100 high-quality original pictures for training.
This model has a very good imitation effect of painting style.
All the training pictures come from the collection of "Ichigo Mashimaro", and the TV version animation picture materials are not mixed in, so as to obtain the pure artist style LORA.
Animagine XL V3 version：https://civarchive.com/models/105656
illustriousXL version：https://civarchive.com/models/889207
One more thing：
Pony Diffusion V6 XL is an excellent base model. It is undeniable that there is a probability of producing good pictures.
However, for anime style, the biggest shortcoming of Pony is that it mixes too many 3D and 2.5D materials, which will lead to the pure 2D style lora. It cannot stably output the specified style, which is even more obvious if there is a character with the same name.
See Ana Coppola's example.
Recommendation Weight：0.6~1.0
Recommendation Model：Pony Diffusion V6 XL