The chapter discusses the role of smaller models versus larger models in the field of AI, highlighting their speed and cost advantages for certain problems. It also explores how fine-tuned smaller models can outperform larger general-purpose models, and predicts the relevance of small and medium-sized models in the near term.