更新v0.7新增tag<h3 id="heading-169">请忽略那些长脖子，高额头，多手以及意义不明的画面!</h3><h3 id="heading-170">Please IGNORE those Loooooong necks, hiiiiiiiiiigh foreheads, multiple hands and equivocal images!</h3>此表格内容仅用于预览Anime Illust Diffusion XL-V0.52~v0.71模型的画风tag效果，测试prompt简单，不代表模型最佳效果，测试结果仅供参考。此表格中图片的生成参数均为：masterpiece,best quality,ultra detailed,by (___________) ,(1girl:1.2),girl walking on street, facing viewer, female focusNegative prompt: aidxlv05_neg,text,low quality,bad anatomy,worst quality,displeasing,ugly,messySteps: 40, Sampler: DPM++ 2M Karras, CFG scale: 7, Size: 1024*1536, Model: animeIllustDiffusion_v052 (少部分为V0.6Alpha） VAE: sdxl_vae.safetensors, Clip skip: 2The contents of this table are only used to preview the effect of Anime Illust Diffusion XL-V5.2 model. The test prompt is simple and does not represent the best effect of the model. The test results are for reference only.The parameters for generating images in this table are：masterpiece,best quality,ultra detailed,by (___________) ,(1girl:1.2),girl walking on street, facing viewer, female focusNegative prompt: aidxlv05_neg,text,low quality,bad anatomy,worst quality,displeasing,ugly,messySteps: 40, Sampler: DPM++ 2M Karras, CFG scale: 7, Size: 1024*1536, Model: animeIllustDiffusion_v052 (A few used V0.6Alpha） VAE: sdxl_vae.safetensors, Clip skip: 2作者：技术实在太菜

Anime Illust Diffusion XL模型全风格Tag测试文档丨Anime Illust Diffusion XL Style Prompt Test

animeIllustDiffusionXLTag_v052.zip

v0.52

v0.7 images

Files with SHA256: e27f8f6dffedacf8e895e010b63e909d0283830fb52735d88ec712bd1925e789