RAG
Retrieval-augmented generation fetches relevant documents at question time and gives them to a model with the question.
What it means
Retrieval-augmented generation, usually shortened to RAG, is a pattern in which a system retrieves relevant documents when a question arrives and passes them to a language model along with the question. The model then writes its answer using that material instead of relying only on what it learned in training. A search API is a common retrieval source, especially when the answer depends on information that changed after training.
RAG does not guarantee accuracy by itself. The model can still misread a passage, ignore one, or state a claim the passage does not support. Good RAG systems keep source links, check quotes against the text, and show users where each claim came from. When you compare search APIs for RAG, look at the passage format, the number of results, latency, and whether the provider returns enough text to verify an answer.
Continue this reading path
Cite this entry
https://apisearchengine.com/glossary/rag/