Vanna AI empowers users to extract insights from their databases simply by asking questions in natural language. It leverages a Retrieval Augmented Generation (RAG) model, trained on your database schema, documentation, and queries, to generate accurate SQL queries. This allows both non-technical users and SQL experts to quickly and easily access data without needing in-depth knowledge of SQL. The core Vanna package is open-source and designed to be highly customizable.
Vanna’s flexible architecture allows users to deploy it within their own infrastructure, using their preferred databases, LLMs, and front-end integrations such as Slack or web applications. Data privacy is a priority; your actual database contents are never sent to the LLM unless explicitly enabled, while the system securely transmits only schemas, documentation, and queries. Vanna offers both a free open-source version and enterprise services to suit various needs from personal use to large-scale deployments, continuously improving as users engage with it, allowing for customized implementations.