Open the gate.. welcome to this dimension..
So more on AI image generation. Not surprisingly since 'stable diffusion' released its models to the world, various folks have been packaging up processing frameworks to make using the models easier.
This one, 'imaganAIry', for a Linux/OSX, builds a wrapper using Python/PyTorch etc. so you can run it locally. Even more surprising (than running fully local) is it'll run on my now somewhat archaic Intel MBPro despite not really having a serious GPU on board - admittedly it takes 15 minutes to generate an image but the fact that it can at all should be classified as a minor miracle 😛
And the final surprise is the quality of the output - they're definitely the best I've seen so far using default settings and very simple text prompts..
Here's some cats and dogs playing poker.. 😃 😛