Trained on SDXL 1.0_vae_fix with an image size of 1024px
Original model size after training. No resizing afterwards.
All example images are raw outputs of the used checkpoint. No additional magic was used.
And yet again, there is still a lot of room for improvement..