Glossary
RAG
RAG
Retrieval-augmented generation: fetching relevant documents and giving them to an AI model before it answers.
RAG (Retrieval-Augmented Generation) is a technique where an app looks up relevant information first, then hands it to an LLM along with the question. The model answers from your data instead of only from what it learned in training.
How it works
- Your documents are split into chunks and turned into embeddings.
- When a user asks something, the app finds the chunks most similar to the question.
- Those chunks are added to the prompt, and the model writes an answer based on them.
Why it matters
RAG lets AI answer questions about your docs, products or knowledge base without retraining a model. It also reduces made-up answers, because the model can point to real sources.
Related: Context Window.