Quickshot

Currently it's not clear which image model is being used when a prompt is submitted. It would be nice to specify which one should be used so we can experiment with different models for the same prompt