---
title: "Teaching a smaller model"
description: "Small language models and how they get good: distillation trains a compact student to imitate a large teacher's outputs (or reasoning traces), capturing mo"
canonical_url: "https://fanout.sh/ai-engineering/lessons/slm-and-distillation"
md_url: "https://fanout.sh/ai-engineering/lessons/slm-and-distillation.md"
access: "public"
---

# Teaching a smaller model

Small language models and how they get good: distillation trains a compact student to imitate a large teacher's outputs (or reasoning traces), capturing mo

## Public overview

Small language models and how they get good: distillation trains a compact student to imitate a large teacher's outputs (or reasoning traces), capturing most of the capability at a fraction of the cost. When a 3B model fine-tuned on your task beats a frontier API on latency, price, and privacy.

---
This representation contains public Fanout content only. Protected Pro lessons, account data, billing, checkout, and pricing are not included.

Browse the public content map: https://fanout.sh/sitemap.md