Chat API (RAG)
AI-powered document Q&A using Retrieval-Augmented Generation.
How it works
- Your question is converted to an embedding vector
- The system searches for relevant document chunks
- Top matching chunks are sent to the LLM with your question
- The LLM generates an answer with source citations