What is Retrieval-Augmented Generation (RAG)?

Retrieval-Augmented Generation (RAG) is an AI architecture where a language model is given access to an external knowledge base or document store. When a query comes in, the system first retrieves the most relevant documents, then passes those documents along with the query to the LLM to generate a grounded, accurate response.

RAG is particularly useful for enterprise applications where you want an AI to answer questions based on your internal documents, knowledge bases, or proprietary data — without the expense and risk of fine-tuning a model or leaking data to a third party.

A concrete example: a procurement company using RAG can have an AI assistant that answers questions about vendor contracts by retrieving the relevant clauses from a document store — rather than hoping the LLM's training data happens to contain the right information.