image to image inference?

Is there any support for image-to-image via the gradio app? Specifically, I see from the diffusers docs at HF that the prior pipeline superclass has an images parameter and describes it as being one or more images that guide the inference.

Will anything useful happen if I pass an image and strength value as parameters to generate_prior in app.py? It sounds too simple to be correct, so I suspect it isn't. If that's the case, what do I actually need to do?

Thanks!