And we can learn a lot from how it is going to win.
in a response to your twitter thread
I go into a sci-fi question, and then drill down into the physics of "how it can be done" - this thing is astounding and probably will end up either now or in a single generation as a college student's best friend in the world - a tutor that can answer any question, forces you to consider how to ask the question in the first place, and then gives you direct, non judgemental, non confrontational information about exactly what you asked with just the smallest amount of spin.. which you can then use to derive further insight.
Well, this turned out to be spot on.
For MacOS the easiest way to use image generation currently is with DiffusionBee. It uses a copy of the Stable Diffusion system running locally on the machine.
I am the non-sophisticated user. I love Dall-E but am finding the more I learn about prompts, the less I am liking my results...