LoRA of Danielle MacDonald for SDXL v1.0.

IMPORTANT: Use trigger word "Maggie Gyllenhaal". I seemed to get much better results by using a real person already known by stable diffusion for the training model instead of using the SDXL base model.

I tried this on many checkpoints and usually get good results on most I've tried. Please reference the .PNG examples for the actual settings used for each. Just drag them into ComfyUI and it'll build the workflow automatically. I used a clip of 2 or 3 most of the time. I didn't use any embeddings, LoRAs, etc.. on any of the examples posted, these came directly from the LoRA and Checkpoint files.

Message me with any requests, pointers, or suggestions. Hope you like this LoRA, if so please like! Thanks!