This model was originally<a rel="ugc" href="https://huggingface.co/minimaxir/sdxl-wrong-lora"> uploaded to Huggingface</a> by<a rel="ugc" href="https://huggingface.co/minimaxir"> Max Woolf </a>- if Max would like to claim ownership of this model, please contact us.A LoRA for SDXL 1.0 Base which improves output image quality after loading it and using <code>wrong</code> as a negative prompt during inference. You can demo image generation using this LoRA in <a target="_blank" rel="ugc" href="https://colab.research.google.com/github/minimaxir/sdxl-experiments/blob/main/sdxl_image_generation.ipynb">this Colab Notebook</a>.Benefits of using this LoRA:<ul><li>Higher detail in textures/fabrics, particularly at full 1024x1024 resolution.</li><li>Higher color saturation and vibrance.</li><li>Higher sharpness for blurry/background objects.</li><li>Better at anatomically-correct hands.</li><li>Less likely to have random artifacts.</li><li>Appears to allow the model to follow the input prompt with a more expected behavior, particularly with prompt weighting such as the <a target="_blank" rel="ugc" href="https://github.com/damian0815/compel">Compel</a> syntax.</li></ul>

Files with SHA256: 1d34fe96f3307d5a0a85cceae56ae160f43b45dae7847d44baa3979ea6ca11cc