Browse Teach Curate

Retrieval-Augmented Generation (RAG)

Grounding LLM outputs in external knowledge: embedding documents, nearest-neighbor retrieval, and augmenting the prompt with retrieved context to reduce hallucination.

Sonny Text

Grounding Model Answers in Your Own Documents with RAG — Building with the Claude API

Retrieval-Augmented Generation retrieves relevant passages and puts them in the prompt so the model answers from your corpus instead of its training data. Understand the retrieve-then-generate flow e…