Based on <a target="_blank" rel="ugc" href="https://huggingface.co/JosephusCheung/ACertainModel">ACertainModel</a>Style from AbyssOrangeMix3. Trained with additional 1 million images, mainly focused on anime bondage, using danbooru style tags.No VAE baked in, style is not very pleasing. Only good for generating some types of bondage and wooden horse. I say it's not worth my dataset and GPU time. Maybe I will improve this on SDXL with a better tagger that I am developing.Our discord: <a target="_blank" rel="ugc" href="https://discord.gg/cdeWjsrgkm">https://discord.gg/cdeWjsrgkm</a>基于<a target="_blank" rel="ugc" href="https://huggingface.co/JosephusCheung/ACertainModel">ACertainModel</a>风格来自AbyssOrangeMix3。使用额外1百万张图训练，主要适用于动画捆绑图，使用danbooru风格提示词。没有VAE，风格不怎么好看。只适合生成某些类型的捆绑和三角木马。对不起这数据集和我的GPU资源。之后可能会使用我在开发的新标签器在SDXL基础上重新训练个。Discord: <a target="_blank" rel="ugc" href="https://discord.gg/cdeWjsrgkm">https://discord.gg/cdeWjsrgkm</a>

Files with SHA256: f14c9e079af1c7b6c9e14b8739ecb1ae7f1416a03ac83df6215278dca5a0b28d