Back to FAQ
General AI FAQ
What is RAG?
RAG means Retrieval-Augmented Generation: the system retrieves relevant knowledge first and then uses it to ground the answer.
RAG is one of the most useful enterprise patterns because it reduces hallucinations and improves relevance by grounding the model in company knowledge.
Good RAG usually combines semantic retrieval, keyword search, metadata filters, permissions and reranking so the model sees the right information for the right user.