这个模型使用了超过300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。训练图片全部来自该画师的原画集。
本人的碎碎念：
这个lora模型并不保证和所有底模都能有很好的兼容性，并不保证所有情况下能生成美图，想获得较好效果请参阅以下推荐。
推荐权重：0.6~1.0
推荐CFG：5~6
推荐模型：Animagine XL V3
--- --- --- --- --- --- --- ---
This model used more than 300 high-quality original pictures for training.
This model has a good effect of restoring the artist's style. The training pictures are all from the original painting collection of the artist.
My thoughts:
This lora model does not guarantee good compatibility with all base models, nor does it guarantee that beautiful pictures can be generated in all situations. If you want to get better results, please refer to the following recommendations.
Recommendation Weight：0.6~1.0
Recommendation CFG：5~6
Recommendation Model：Animagine XL V3
Description
FAQ
Comments (3)
works with pony diffusion?
As you know, PONY XL is a great base model, and I have also used it to make lora models. However, the current test results are not as good as Animagine XL V3, especially those with strong 2D style animation characters.
I will continue to pay attention. If the subsequent updated model has better results, I will also simultaneously produce a lora model based on PONY XL.
Thank you very much for your testing and support~
@flyx3 In my experience, it always gives excellent results with Pony checkpoint models. How is this possible? What is true, a 'style' is not guaranteed, but whatever result appears from the combination of this SDXL LoRA and a Pony checkpoint model, it remains consistent in all trials. She strikes like a supermodel who can actually act what's very fitting to Touhou mythology.