Glossary
What is RAG?
Retrieval Augmented Generation for grounded AI responses.
What is RAG?
RAG (Retrieval Augmented Generation) is a technique that retrieves relevant documents from a knowledge base and includes them in the LLM prompt. This grounds the model's response in factual, up-to-date information rather than relying solely on training data.
How RAG Works
- 1. Query: User asks a question
- 2. Retrieve: Search vector database for relevant docs
- 3. Augment: Add retrieved docs to prompt
- 4. Generate: LLM responds using retrieved context
RAG Benefits
- Reduces hallucinations with factual grounding
- Enables up-to-date information
- Provides source attribution
- Works with proprietary data
RAG Limitations
- Retrieval quality affects output quality
- Models can still misinterpret sources
- Can hallucinate connections between facts
- Requires monitoring for accuracy
Does RAG eliminate hallucinations?
No. RAG significantly reduces hallucinations but doesn't eliminate them. Models can still misinterpret retrieved documents, hallucinate connections, or generate claims not supported by sources. Monitor RAG outputs for accuracy.
Monitor RAG pipeline quality
Start Free