An AWS Show and Tell session on running Llama 4 Scout and Maverick on AWS — model choice, deployment shape and cost tradeoffs for teams considering open-weight foundations.