<h1 id="oeai-fine-tuned-model-based-on-illustrious-xl-1.1-uugq1pjwm">OeAi - Fine-tuned model based on illustrious XL 1.1</h1><ul><li>Slightly tweaked illustrious XL 1.1 to be more flat and light-coloerd</li><li>Even without using any danbooru style prompts, it can still perform well in image generation</li><li>Added danbooru tag set style artist name to replicate the style</li><li>You can use it like NAI and train it as a base model</li><li>Using the danbooru tag set prompts has a better effect than illustrious XL 1.1 (better anime illustration style)</li><li>About image generation capabilities:This model works well with high-resolution generated images, so it doesn’t need subsampling (the native resolution is 1536x1536, see my example images for details, they all come with comfyui workflow information)</li><li>set up:Sampling steps: 26CFG: 7Sampler: euler_ancestralScheduling: ddim_uniformResolution: 1536x1536, 1536x2112, 1472x2240, 1664x1984(To generate horizontal images, reverse the recommended resolution width and height)</li></ul><h1 id="oeai-illustrious-xl-1.1-cnnlw65s9">OeAi - 基于illustrious XL 1.1的微调模型</h1><ul><li>将illustrious XL 1.1微调至更偏向平涂浅色型风格</li></ul><ul><li>即使是不使用任何danbooru风格提示词它对图像的生成能力上也能表现良好</li><li>添加danbooru标签集风格的画师名来复刻风格</li><li>你可以将它当作NAI一样使用并作为基本模型进行训练</li><li>使用danbooru标签集提示词相较illustrious XL 1.1有着更好的效果（更好的动漫插画风格）</li><li>关于图像生成能力：此模型在使用高分辨率生成图像时依旧表现良好，因此它无需二次采样（原始分辨率支持1536x1536，具体可以查看我的示例图片，它们都带有comfyui的工作流信息）</li><li>参数：采样步数：26CFG：7采样器：euler_ancestral调度：ddim_uniform分辨率：1536x1536，1536x2112，1472x2240，1664x1984（生成横图将推荐分辨率宽高反过来）</li></ul>

OeAi - illustrious XL (More stunning illustration style or cute style)

OeAi v1.2-UE is based on v1.0 version and fine-tuned with a training set similar to v1.1-U version. The training set has been greatly modified and parameters have been added, making v1.2-UE version more perfect!*The hand details are better than v1.1-U version*The color is lighter and softer than v1.1-U version, and the overall picture is more comfortable*The watermark issue has been fixed*Better in illustration tasks!OeAi v1.2-UE是基于v1.0版本使用v1.1-U版本相似的训练集进行重新微调，对训练集进行了大幅修改并增加参数，使得v1.2-UE版本更加完美！*手部细节相对与v1.1-U版本更好*色彩上对比v1.1-U版本更浅更柔和，整体画面更加舒适*对水印的问题得到了修复*在插画任务上更加出色！

oeaiIllustriousXLMore_oeaiV12UE.safetensors

OeAi v1.2-UE

OeAi v1.1-U

OeAi v1.0

OeAi - Fine-tuned model based on illustrious XL 1.1<ul><li>Slightly tweaked illustrious XL 1.1 to be more flat and light-coloerd</li><li>Even without using any danbooru style prompts, it can still perform well in image generation</li><li>Added danbooru tag set style artist name to replicate the style</li><li>You can use it like NAI and train it as a base model</li><li>Using the danbooru tag set prompts has a better effect than illustrious XL 1.1 (better anime illustration style)</li><li>About image generation capabilities:This model works well with high-resolution generated images, so it doesn’t need subsampling (the native resolution is 1536x1536, see my example images for details, they all come with comfyui workflow information)</li><li>set up:Sampling steps: 26CFG: 7Sampler: euler_ancestralScheduling: ddim_uniformOeAi - 基于illustrious XL 1.1的微调模型</li></ul><ul><li>将illustrious XL 1.1微调至更偏向平涂浅色型风格</li></ul><ul><li>即使是不使用任何danbooru风格提示词它对图像的生成能力上也能表现良好</li><li>添加danbooru标签集风格的画师名来复刻风格</li><li>你可以将它当作NAI一样使用并作为基本模型进行训练</li><li>使用danbooru标签集提示词相较illustrious XL 1.1有着更好的效果（更好的动漫插画风格）</li><li>关于图像生成能力：此模型在使用高分辨率生成图像时依旧表现良好，因此它无需二次采样（原始分辨率支持1536x1536，具体可以查看我的示例图片，它们都带有comfyui的工作流信息）</li><li>参数：采样步数：26CFG：7采样器：euler_ancestral调度：ddim_uniform</li></ul>

OeAi - illustrious XL

OeAi v1.2-UE is based on v1.0 version and fine-tuned with a training set similar to v1.1-U version. The training set has been greatly modified and parameters have been added, making v1.2-UE version more perfect!
*The hand details are better than v1.1-U version
*The color is lighter and softer than v1.1-U version, and the overall picture is more comfortable
*Fixed watermark issue
*Better in illustration tasks!

OeAi v1.2-UE是基于v1.0版本使用v1.1-U版本相似的训练集进行重新微调，对训练集进行了大幅修改并增加参数，使得v1.2-UE版本更加完美！
*手部细节相对与v1.1-U版本更好
*色彩上对比v1.1-U版本更浅更柔和，整体画面更加舒适
*修复了水印问题
*在插画任务上更加出色！

Files with SHA256: 0de8ce19410ee5f8175c2d6239fbfda5087884a1e156debc66ba03e9eece24ff