Retrieval-augmented generation
Also known as: RAG
What is retrieval-augmented generation?
Retrieval-augmented generation is a pattern where relevant documents are fetched from a knowledge base at query time and inserted into the model's prompt, so answers are grounded in your own content rather than the model's training data.
Why it matters
RAG is the standard way to make a general model answer accurately about a specific business without fine-tuning. It also makes answers citable, which matters for support, compliance and internal search.
Where it fails
- Bad chunking — retrieved passages lack the context needed to answer
- No evaluation set, so quality regressions go unnoticed
- Stale index after source content changes
Last reviewed by the gAIcko editorial team.