Beating The DALL-E Mini: Too much traffic, please try again

Dall-E seems only aware of one pyramid. But anyway, this is pretty interesting. I only have to press the button once and sometimes twice to make it work. Maybe it's the time of day, or everyone from one (or twenty) sites hitting it at the same time. It doesn't seem to need a script or any fiddly bits to make it work.

I thought this would be about best practices for deployment of models like this.

I once asked in the LAION discord what that was and they said basically flask/fastapi server + load the model + gunicorn + ec2 instances.

I'm curious if any HN readers have any further insight or thoughts on ML model deployment for projects like DALL-E Mini.