ScriptsMar 31, 2026·2 min read

Vanna — Chat with Your SQL Database Using AI

AI-powered text-to-SQL. Ask questions in plain English, get accurate SQL queries and visualizations. Learns your schema. Works with any database. 23K+ stars.

TO
TokRepo精选 · Community
Quick Use

Use it first, then decide how deep to go

This block should tell both the user and the agent what to copy, install, and apply first.

pip install vanna
import vanna
from vanna.openai import OpenAI_Chat
from vanna.chromadb import ChromaDB_VectorStore

class MyVanna(ChromaDB_VectorStore, OpenAI_Chat):
    def __init__(self, config=None):
        ChromaDB_VectorStore.__init__(self, config=config)
        OpenAI_Chat.__init__(self, config=config)

vn = MyVanna(config={'api_key': 'sk-...', 'model': 'gpt-4o'})
vn.connect_to_postgres(host='localhost', dbname='mydb', user='postgres', password='pass')

# Train on your schema
vn.train(ddl="CREATE TABLE users (id INT, name TEXT, email TEXT, created_at TIMESTAMP)")

# Ask questions
sql = vn.generate_sql("How many users signed up this month?")
print(sql)  # SELECT COUNT(*) FROM users WHERE created_at >= '2026-03-01'

df = vn.run_sql(sql)
vn.generate_plotly(question="...", sql=sql, df=df)  # Auto-visualization
Intro

Vanna is an AI-powered text-to-SQL tool. Ask questions about your database in plain English and get accurate SQL queries, results, and auto-generated visualizations. It learns your specific schema, naming conventions, and business logic through training — getting more accurate over time. Works with PostgreSQL, MySQL, SQLite, BigQuery, Snowflake, and more. 23,000+ GitHub stars, MIT licensed.

Best for: Data analysts and developers who want natural language access to databases Works with: PostgreSQL, MySQL, SQLite, BigQuery, Snowflake, DuckDB, any SQL database

Key Features

Agentic RAG for SQL

Uses retrieval-augmented generation specifically tuned for SQL. Retrieves relevant schema, past queries, and documentation before generating SQL.

Schema Learning

Train Vanna on your DDL, sample queries, and documentation. It learns your tables, relationships, and naming patterns.

Auto-Visualization

Generates Plotly charts automatically from query results — bar charts, line graphs, pie charts, and more.

Web UI

Built-in Flask web interface for non-technical users to query databases:

from vanna.flask import VannaFlaskApp
app = VannaFlaskApp(vn)
app.run()

Multi-Database

Connect to PostgreSQL, MySQL, SQLite, BigQuery, Snowflake, DuckDB, MSSQL, and any ODBC database.

FAQ

Q: What is Vanna? A: An AI text-to-SQL tool that converts natural language questions into accurate SQL queries with auto-visualization. Learns your schema over time. 23K+ stars.

Q: How accurate is the SQL generation? A: Accuracy improves as you train Vanna on your schema and example queries. With good training data, 85-95% accuracy on typical business queries.

🙏

Source & Thanks

Created by Vanna AI. Licensed under MIT. vanna-ai/vanna — 23,000+ GitHub stars

Related Assets

ScrapeGraphAI — AI-Powered Web Scraping

Python scraping library powered by LLMs. Describe what you want to extract in natural language, get structured data back. Handles dynamic pages. 23K+ stars.

TokRepo精选

Kotaemon — Open-Source RAG Document Chat

Clean, open-source RAG tool for chatting with your documents. Supports PDF, DOCX, web pages. Multi-model, citation, and multi-user. Self-hostable. 25K+ stars.

TokRepo精选

Dagger — Programmable CI/CD Engine

Run CI/CD pipelines as code — locally, in CI, or in the cloud. Replace YAML with real programming languages. Cacheable, portable, testable. 15.6K+ stars.

TokRepo精选
◈Home🏆Trending👤Me