Skip to content
Velaris

Technique

What is Retrieval-Augmented Generation?

Also known as: RAG

Retrieval-augmented generation (RAG) improves an AI's answers by first retrieving relevant documents and feeding them to the model as context.

Instead of relying only on what a model memorized in training, RAG fetches relevant, up-to-date information at query time — from your files, a knowledge base, or the web — and grounds the answer in it.

This reduces hallucination and lets an assistant answer from your private data. It's a core technique behind research assistants and any agent that needs to reason over specific documents.

Related terms

Vector Search Hallucination Large Language Model

See it in action

Velaris puts retrieval-augmented generation to work inside a real AI Operating System.

Get early access