Overview of Small Language Models (SLMs) in solving business problems and improving device capabilities. Compares SLM with Large Language Models (LLMs) and explains why SLM may be better suited for an organization's needs, with emphasis on customization via fine-tuning. Covers deploying models on edge devices for real-time processing and decision-making, plus a roadmap from cloud development to cross-platform deployment, with Azure for seamless integration and IP ownership. Concludes with how Microsoft's Student Ambassador Program can empower students to bridge academia and industry and foster innovation.

We will dive deeper into inference from Apple Silicon to huge production clusters, and cover tricks how to make it faster.