---
title: "RAG | Glossary"
description: RAG, retrieval-augmented generation, means the model must fetch named passages before it answers. The retrieval is the product. The generation is the last…
image: "https://lenvanderhof.com/photos/len-portrait-1.jpg"
---

Term

# RAG

RAG, retrieval-augmented generation, means the model must fetch named passages before it answers. The retrieval is the product. The generation is the last mile. If you cannot inspect which chunk was used, you do not have RAG. You have a language model with extra context stuffed in the prompt. A vector store is a component. GRAIN is the loop that keeps the answer honest. Adding embeddings is not a finished system. Naming the passage is.

## When to use it

Use it when a team is treating “we added a vector store” as a finished system. If nobody can open the chunk behind the sentence, you shipped generation, not retrieval.

## Do not confuse with

Not “the model read our docs.” RAG is fetch-then-generate with a named passage. GRAIN is the inspectable loop. The RAG Engineer is available now in English and Dutch paperback, hardcover, and Kindle.

## Essays on this site

- [What is RAG? Retrieval, then generation](https://lenvanderhof.com/en/blog/what-is-rag/)

```json
{
	"@context": "https://schema.org",
	"@graph": [
		{
			"@type": "DefinedTerm",
			"@id": "https://lenvanderhof.com/glossary/rag/#term",
			"name": "RAG",
			"description": "RAG, retrieval-augmented generation, means the model must fetch named passages before it answers. The retrieval is the product. The generation is the last mile. If you cannot inspect which chunk was used, you do not have RAG. You have a language model with extra context stuffed in the prompt. A vector store is a component. GRAIN is the loop that keeps the answer honest. Adding embeddings is not a finished system. Naming the passage is.",
			"url": "https://lenvanderhof.com/glossary/rag/",
			"inDefinedTermSet": "https://lenvanderhof.com/glossary/"
		},
		{
			"@type": "WebPage",
			"@id": "https://lenvanderhof.com/glossary/rag/#webpage",
			"url": "https://lenvanderhof.com/glossary/rag/",
			"name": "RAG",
			"description": "RAG, retrieval-augmented generation, means the model must fetch named passages before it answers. The retrieval is the product. The generation is the last mile. If you cannot inspect which chunk was used, you do not have RAG. You have a language model with extra context stuffed in the prompt. A vector store is a component. GRAIN is the loop that keeps the answer honest. Adding embeddings is not a finished system. Naming the passage is.",
			"inLanguage": "en",
			"isPartOf": {
				"@id": "https://lenvanderhof.com/#website"
			},
			"about": {
				"@id": "https://lenvanderhof.com/glossary/rag/#term"
			},
			"mainEntity": {
				"@id": "https://lenvanderhof.com/glossary/rag/#term"
			},
			"breadcrumb": {
				"@id": "https://lenvanderhof.com/glossary/rag/#breadcrumb"
			}
		},
		{
			"@type": "BreadcrumbList",
			"@id": "https://lenvanderhof.com/glossary/rag/#breadcrumb",
			"itemListElement": [
				{
					"@type": "ListItem",
					"position": 1,
					"name": "Home",
					"item": "https://lenvanderhof.com/"
				},
				{
					"@type": "ListItem",
					"position": 2,
					"name": "All terms",
					"item": "https://lenvanderhof.com/glossary/"
				},
				{
					"@type": "ListItem",
					"position": 3,
					"name": "RAG",
					"item": "https://lenvanderhof.com/glossary/rag/"
				}
			]
		}
	]
}
```