# Retrieval-Augmented Generation (RAG)

> Retrieval-augmented generation is the pattern where an AI model fetches relevant material from a knowledge source at question time and answers from what it retrieved, rather than from its training data alone. The model provides the reasoning; the retrieval provides the facts.

- Part of: [The Sentry AI Transformation Glossary](https://sentrysolutions.ai/glossary)
- Canonical: https://sentrysolutions.ai/glossary/retrieval-augmented-generation

RAG is how a general model answers with your business's specifics without being retrained on them. The quality ceiling, though, is set by the retrieval, not the model: if the search returns the wrong five paragraphs, the best model in the world summarises the wrong five paragraphs confidently.

That is why RAG quality work is mostly knowledge-base work: layout, chunking, embeddings and structure, so retrieval answers in one hop instead of five. A knowledge graph raises the ceiling further by giving retrieval relationships to follow rather than just text to match.

## Related terms

- [Company Knowledge Graph](https://sentrysolutions.ai/glossary/company-knowledge-graph)
- [Context Engineering](https://sentrysolutions.ai/glossary/context-engineering)
- [Fine-Tuned Models](https://sentrysolutions.ai/glossary/custom-fine-tuned-models)