Help fuel my passion! Even $1 makes a difference.
这是一个全新的精调训练lora模型，使用超过200张高质量图片进行训练。
训练图片全部来自该画师的原画集，得到纯净的小岛文美风格LORA。
推荐权重：0.8~1.5
Animagine XL和 SD1.5 版本：https://civarchive.com/models/223383
--- --- --- --- --- --- --- ---
This is a brand new fine-tuned training lora model, which is trained with more than 200 high-quality pictures.
All the training pictures are from the original painting collection of the artist,, so that the pure Ayami Kojima style lora is obtained.
Recommendation Weight：0.8~1.5
Animagine XL and SD1.5 version：https://civarchive.com/models/223383
Description
FAQ
Comments (8)
实在太厉害了！话说大佬，训练风格Lora最少需要多少图片才合适呢？总步数控制在多少比较好？
Awesome style!
wow I don't know this artist but it's beautiful, and the unusual thing is this time I like ALL the examples you posted. I'm gonna show this page to anyone who tells me AI is bad. I mean look at this !!
Thanks for sharing
Thanks for the support, but AI still can't replace humans~
恶魔城
"WHAT IS A MAN?"