Natural Language to SQL

Database First.

Convert plain English questions into safe, optimized SQL queries. Powered by 100+ LLM providers with enterprise-grade security.

Get Started View on GitHub
CLI REST API Python PHP Node.js Go
SmartQL Interactive Shell
Enter natural language questions to generate SQL.
Commands: /help, /schema, /execute, /clear, /quit
> /execute
Auto-execute mode: ON
> Show me top 5 customers by total spending
SELECT c.name, SUM(o.total) AS spending
FROM customers c
JOIN orders o ON o.customer_id = c.id
GROUP BY c.id
ORDER BY spending DESC LIMIT 5
Results: 5 rows
{'name': 'Acme Corp', 'spending': 24500.00}
{'name': 'TechStart Inc', 'spending': 18200.00}
{'name': 'Global Trade', 'spending': 15800.00}
{'name': 'DataFlow LLC', 'spending': 12400.00}
{'name': 'CloudNine', 'spending': 9850.00}
> _

Why SmartQL?

🤖

100+ LLM Providers

OpenAI, Claude, Gemini, Groq, Ollama, and more via LiteLLM. Switch providers with one line.

🔒

Enterprise Security

AST-based SQL validation with sqlglot. Table allowlists, column blocking, query limits.

📊

Semantic Layer

YAML-based configuration with business context, aliases, and rules. Your database, your language.

🔍

Auto-Introspection

Automatically discover your database schema. Works out of the box with zero configuration.

⚡

Fast & Local

Run with local models via Ollama. No data leaves your infrastructure.

🛠️

CLI & API

Interactive shell for testing. REST API for integration. Python library for embedding.

How It Works

1

Connect Your Database

database:
  type: mysql
  connection:
    host: localhost
    database: myapp
2

Choose Your LLM

llm:
  provider: groq
  groq:
    model: llama-3.1-8b-instant
3

Ask Questions

from smartql import SmartQL

sql = SmartQL.from_yaml("config.yml")
result = sql.ask("Top 10 customers by revenue")

Works With Your Stack

PostgreSQL MySQL SQLite SQL Server

Ready to Query in Plain English?

Get started in minutes. Open source and free forever.

Get Started Star on GitHub