Retrieval-augmented generation (RAG)

Retrieval-Augmented Generation is an approach that combines a retrieval system with a generative AI model to produce more accurate and fact-based responses. Instead of relying only on stored training data, RAG retrieves relevant information from external sources during response generation.

In a RAG system, the model first searches a knowledge base or database for related content and then uses that information to create its answer. This approach improves accuracy, reduces outdated outputs, and helps maintain factual consistency.

RAG is widely used with large language models that need up-to-date or domain-specific information. It’s especially valuable in research, customer support, and enterprise applications that depend on current and verified data.