OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Data Agents

Vanna AI

Vanna is an open-source AI SQL agent framework for building natural-language interfaces to organizational databases and systems of record. It supports multi-turn conversations, configurable language-model providers and databases, and access-controlled query execution.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Vanna AI — OpenViST project profile in Data Agents.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Vanna when building an AI assistant that translates conversational questions into database queries and can be configured for different model providers, databases, and access-control needs.

Who it is for

  • Teams building conversational access to organizational databases.
  • Developers creating data assistants for people with varying SQL expertise.
  • Builders who need user-scoped, access-controlled database query experiences.

Documented uses

  • Create a natural-language interface for querying company databases.
  • Build a multi-turn SQL assistant that can handle follow-up questions.
  • Provide database queries through an access-controlled user experience.
Before you choose

Requirements & limits

The homepage offers a sample-data experience marked as requiring no sign-in. Vanna is identified as open source, and its pricing page describes annual subscriptions and tier upgrades.

  • The framework supports different LLM providers and databases, so builders can configure the agent around their chosen components.
  • Multi-turn interaction supports follow-up questions rather than only single-query exchanges.
  • Access-controlled execution is relevant for applications that need to scope database queries by user.
  • The no-sign-in note applies to the sample-data trial, not necessarily to other service actions.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Try Vanna with the Chinook sample database in Google Colab

Explore natural-language queries against Vanna’s sample music-store database.

Before you start

  • Access to Google Colab.
  • The Vanna quick-start sample notebook, which includes the Chinook database.

Steps

  1. Open the Vanna quick-start page.

  2. Open the sample notebook in Google Colab.

  3. Try natural-language questions about the sample data, such as asking for the top-selling albums or sales by country.

Expected outcome

A chance to try natural-language queries against the Chinook sample database in the notebook.

Guide limitations

  • This quick start uses sample data; it does not document connecting to your own database or deploying an agent.
Documentation 1 ↗
Evidence and transparency

Sources & review

vanna.ai ↗Guide documentation ↗
Source record details
Source availability
Active
Sources checked
7 October 2026
Original website title
Vanna 2.0 – Build Agents Your Users Can Actually Use · Official homepage title · retained in its source language
How we review sourcesSuggest a correction