Concept· 2 min
What is RAG?
Retrieval-Augmented Generation lets an LLM look things up before answering, grounding it in your own data.
Key idea
Instead of relying only on the model's memory, RAG retrieves relevant passages from a knowledge base and passes them into the prompt.
Example
A support bot searches your help center for the 3 most relevant articles, then asks the LLM to answer using only those passages — with citations.
What to remember
- Reduces hallucinations by grounding answers in real sources
- Cheaper than fine-tuning for keeping knowledge fresh
- Quality depends on your retriever (embeddings + chunking)