Train and deploy Stable Diffusion using AWS Trainium and Inferentia

Are you developing generative AI models and in need of high performance while keeping your costs low?

AWS Trainium and AWS Inferentia are ML accelerators purpose-built for large-scale deep learning. Learn about a complete solution for training and deploying a text-to-image diffusion model. We demonstrate how to easily compile the model, train it on Trainium, and deploy it on Inferentia. We also look at benchmarks that can help you compare the price and performance of these accelerators with other comparable Amazon EC2 instances. Finally, we give a live demonstration of text-to-image generation working on Trainium and Inferentia.