Catalog

While large language models (LLMs) offer versatility, smaller language models (SLMs) provide resource efficiency, speed, and simplicity. This builders’ session explores task simplification and decomposition techniques to harness multiple specialized SLMs, surpassing a single large model’s accuracy at a fraction of the cost. Learn how to build a generative AI application by deploying SLMs from Amazon SageMaker JumpStart on AWS Inferentia. Understand model trade-offs and a mental model for selecting the right approach. Leave with a deep understanding of designing generative AI applications by orchestrating SLMs through agentic workflows for cost savings, accuracy, and performance. You must bring your laptop to participate.