Definition
Retrieval-Augmented GenerationRAG
A technique that lets a language model answer from your own documents by retrieving relevant passages first and generating the answer from them.
Large language models only know what was in their training data. Retrieval-Augmented Generation (RAG) closes that gap at question time: the system searches a knowledge base for passages relevant to the question, then gives those passages to the model as context and asks it to answer from them.
Because the knowledge lives in a searchable index rather than in the model's weights, RAG systems can be updated instantly, can cite their sources, and can respect access rules about who may see which document. That makes RAG the default architecture for assistants over company documents, legal archives, policies and support content.
The quality of a RAG system depends mostly on retrieval: how documents are parsed and chunked, how search combines keywords and semantics, and whether the result is re-ranked. Evaluating retrieval on real questions is what turns a demo into a production system.
Related services
Related terms