<p>This is the SDXL version of <a target="_blank" rel="ugc" href="https://civarchive.com/models/250314/female-surgeon-surgical-nurse-taking-off-mask">https://civarchive.com/models/250314/female-surgeon-surgical-nurse-taking-off-mask</a>, though it is trained with a largely different dataset. It is pretty hard to find high resolution training data for this concept for SDXL, so I ended up used the SD1.5 version to generate about half of the dataset. </p><p>The LoRA is trained on SDXL base model and should work on both realistic and anime SDXL models. The model has a tendency to draw pocket / nametag on the surgical outfits. You can workaround it by adding these to the negative prompts.</p>

Files with SHA256: 480b22da6b48e344fb8f250e0cd7cf194b94cd98df2d3a9cf73b2ab8576aa306