Melissa Roxburgh LoRA trained using RealisticVision3.0.

I found the best weight for the model is 0.6-0.8.

Model was trained on 768x768.

Some checkpoints are better than others from testing. Feel free to experiment.

Enjoy <3