v7

这是一个可以使用画师名字来调整画风以及只需输入角色名字就可以得到相应角色的模型,接近于NAI3,但是不同于NAI3的是它不但可以使用基于illustrious训练的Lora,还可以使用很大部分SDXL和pony的Lora.它比nai3好的地方就是它初始的画风就足够你使用,而且拥有良好的肢体跟手脚.而且也不需要很复杂的tag就能得到很好的图像.This is a model that allows you to adjust the art style by using the artist's name and generate corresponding character models by simply inputting the character's name. It is similar to NAI3, but unlike NAI3, it not only supports LoRA trained on Illustrious but also works with a large portion of SDXL and Pony LoRA.What makes it better than NAI3 is that its initial art style is already good enough for use, with well-defined body proportions and hands/feet. Additionally, it doesn't require complex tags to generate high-quality images.画师可以参考 https://docs.qq.com/sheet/DZGxRSXhvcmNmeHFv?u=32e2cc2b814144e6b534a1756abdeed5&tab=9v7zie Recommended settings:Steps: 30CFG scale: 5-7Sampler: Euler aNot very necessary to use ADetailer correctionPositive Promptmasterpiece,best quality,amazing quality,very aesthetic,absurdres,newest,Negative Promptlowres,(bad quality,worst quality:1.2),bad anatomy,sketch,jpeg artifacts,ugly, poorly drawn,(censor),blurry,watermark,simple background,transparent background