Type | |
Stats | 5,175 34,283 |
Reviews | |
Published | Mar 15, 2024 |
Base Model | |
Training | Steps: 23,800 Epochs: 15 |
Usage Tips | Clip Skip: 2 |
Trigger Words | souryuu asuka langley ayanami rei katsuragi misato ikari shinji nagisa kaworu nadia la arwall |
Hash | AutoV2 D3C88A44BF |
这个模型使用了超过270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。
FLUX版本:https://civitai.com/models/708851
Animagine XL V3版本:https://civitai.com/models/265526
illustriousXL 版本:https://civitai.com/models/406695
SD1.5版本:https://civitai.com/models/114608
碎碎念:
Pony Diffusion V6 XL是一个优秀的底模,但是对于二次元动漫风格(2D),最大的缺点就是并不能稳定的输出指定风格,并且在色彩方面始终有一些写实。
碎碎念2:(仅中文)
其实训练了一堆PONY的lora,但一直没发布(毕竟训练集和打标和A3都是一样的,纯换个底模罢了),一直有人求,想了想还是发了吧~~嗯,就是这样~ ╮(╯_╰)╭
--- --- --- --- --- --- --- ---
This model used more than 270 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.
FLUX version:https://civitai.com/models/708851
Animagine XL V3 version:https://civitai.com/models/265526
illustriousXL version:https://civitai.com/models/406695
SD1.5:https://civitai.com/models/114608
One more thing:
Pony Diffusion V6 XL is an excellent base model, but for the two-dimensional animation style (2D), the biggest disadvantage is that it cannot stably output the specified style, and it is always somewhat realistic in terms of color.