There are a variety of batch use cases where it would be great to integrate Photo AI, but where it’s currently extremely expensive. These are cases where you want Photo AI as a step in a processing chain that is part of a batch job. You can do this with the CLI currently, but because each call to the CLI has to load the world, it’s very slow even on very high end hardware. I have a beast of a machine and it’s like 14 seconds each I think. It would be great if you could avoid this via some kind of client/server model or python interface. Even something as simple as, if Photo AI is already open, the CLI just passes the image to it to process so that you don’t have to load everything up every time.
