What is retrieval-augmented generation (RAG)?
Also called: RAG, retrieval augmented generation
Retrieval-augmented generation (RAG) is a way to make an AI answer from specific, trusted documents. Before the model answers, the system searches your documents for the most relevant passages and gives them to the model, which then answers using, and ideally citing, those sources.
How does retrieval-augmented generation work?
Documents are split into chunks and turned into embeddings stored in a vector database. A question is embedded too, the closest chunks are retrieved, and they're added to the prompt with an instruction to answer only from them.
Why does retrieval-augmented generation matter?
RAG reduces hallucinations and lets AI answer about your own data, like company policies or course notes, without retraining the model.
An example
Example: Plintrix's tutor, Pli, uses RAG: it finds the relevant lesson passages and cites them in its answer.
Related terms
Learn it properly
Retrieval-augmented generation is covered in AI Builder, with short lessons, practice questions and instant feedback.