📚 Glossary
RAG (Retrieval-Augmented Generation)
In one line: A technique for grounding AI answers in your own documents — retrieve relevant context first, then generate the answer.
Retrieval-Augmented Generation (RAG) is the technique of giving an LLM access to a private knowledge base. The flow:
- User asks a question.
- Retrieve the most relevant chunks from a knowledge base using embeddings.
- Generate the answer using the retrieved chunks as context.
RAG is how chatbots can answer questions about your company's internal docs, a specific PDF, or any knowledge that's not in the LLM's training data. It's also a way to mitigate hallucinations — answers are grounded in retrieved sources you can cite.
RAG is conceptually simple but production-ready RAG (chunking, embedding choice, re-ranking, evaluation) is a meaningful engineering effort.
See it in action — ask any AI about rag (retrieval-augmented generation) on AskAI.free.
Try it free →