I would like to run open source public models from Github on AWS Sagemaker serverless inference (eg. https://github.com/ai-forever/Kandinsky-3). I would like it to run on AWS Sagemaker serverless because I would like to API in when needed and only pay for the inference. The problem is I don’t know where to start. I ask 3 different AIs, got 3 different answers, and no idea which one is right.
Could somebody point me to the right direction? Any hint, tutorial, general list of steps, etc.?