感谢B站up@鳳鳴櫻弦or大妮可为本次lora训练贡献的训练集，总共运用了625张唯美画风的数据集，大多数来自midjourney的niji7，你可以轻松的生成大光圈高漫反射的唯美风格图像。

Thank you to Bilibili UP @鳳鳴櫻弦or大妮可 for contributing the training set for this LoRA training. A total of 625 datasets with a beautiful artistic style were used, most of which came from midjourney's niji7. You can easily generate images with a beautiful style featuring a large aperture and high glossiness.

建议lora强度：0.5~0.7

Recommended LoRa strength:0.5~0.7

在使用本lora的时候建议加上一些增强光影的提示词：chiaroscuro, natural light, mood lighting, cinematic lighting, volumetric lighting, ray tracing, dramatic shadows,dappled sunlight, light rays, Lens Flare, overexposure, blurry foreground, depth of field, blurry background,

When using this LORA, it is recommended to add some words that enhance the description of light and shadow: chiaroscuro, natural light, mood lighting, cinematic lighting, volumetric lighting, ray tracing, dramatic shadows,dappled sunlight, light rays, Lens Flare, overexposure, blurry foreground, depth of field, blurry background,

本lora以indigoVoidFurryFusedXL_noobaiV29Variant为底模训练而成，在不同底模上面会有不同的风格效果。

This LoRA was trained based on the indigoVoidFurryFusedXL_noobaiV29Variant as the base model. Different base models will result in different style effects.

如果你有好的建议或者改进意见可以通过邮箱联系我：[email protected]

If you have any good suggestions or improvement opinions, you can contact me via email: [email protected].