Retrieval-augmented generation (RAG)

Retrieval-augmented generation lets an AI model answer from your own documents: relevant passages are searched and retrieved first, then given to the model together with the question.

RAG reduces made-up answers and lets answers cite their sources, without retraining the model. It typically uses a vector database to find passages by meaning rather than exact words.

Examples tracked on AI Stats Live

Retrieval-augmented generation (RAG): example tools with company, category, price and popularity rank
Tool	Company	Category	From	Free tier	Popularity
Model Context Protocol	Anthropic	Agent frameworks	—	Unknown	#1
LangChain	LangChain	Orchestration & RAG	—	Unknown	#10
Cohere Platform	Cohere	Model APIs	—	Yes	#28
NotebookLM	Google	Research	$19.99/mo	Yes	#45
Vercel AI SDK	Vercel	Orchestration & RAG	—	Unknown	#49
Qdrant	Qdrant	Vector databases	—	Unknown	#66
Weaviate	Weaviate	Vector databases	—	Unknown	#69
Chroma	Chroma	Vector databases	—	Unknown	#73
Pinecone	Pinecone	Vector databases	—	Unknown	#75
LlamaIndex	LlamaIndex	Orchestration & RAG	—	Unknown	#77

Popularity is measured attention (Wikipedia, Hacker News, package downloads), not quality. Prices are the cheapest paid individual plan; each profile shows whether the price is verified. Data refreshed 2h ago.

Practical guides

Ask questions about a PDF

Browse categories

Related terms

Large language models (LLMs)

Explanation written by AI Stats Live editors; last reviewed 29 Sep 2026. Examples and rankings update automatically from the sources listed on the methodology page.