# Retrieval-Augmented Generation

> A pattern that retrieves external knowledge and adds it to the prompt before generation.

- Category: Retrieval & Grounding
- Stage: Established
- Also called: RAG
- Canonical: https://www.joelcsaji.com/ai-dictionary/retrieval-augmented-generation

## Plain English

The model looks things up before answering.

## Example

A support assistant retrieves product docs and uses them to answer a customer question with citations.

## Why it matters

RAG is the default way to make LLMs useful with private, current, or domain-specific information.

## Related

- [Grounding](https://www.joelcsaji.com/ai-dictionary/grounding)
- [Embedding](https://www.joelcsaji.com/ai-dictionary/embedding)
- [Vector Database](https://www.joelcsaji.com/ai-dictionary/vector-database)

## Sources

- [Google Cloud Generative AI Glossary](https://docs.cloud.google.com/docs/generative-ai/glossary)
- [IBM: Retrieval-Augmented Generation](https://www.ibm.com/think/topics/retrieval-augmented-generation)