Getting started | Database

SurrealDB stores relational, document, graph, time-series, vector and full-text data in one engine, queried through SurrealQL and reachable from ten official SDKs. It runs embedded in your application, as a single node, or as a distributed cluster.

Already using an AI agent?

Paste one prompt and your agent installs SurrealDB's skills and connects to the MCP server, ready to write SurrealQL against a real database.

Learn more about SurrealDB for agents

Where do you want to start?

I am new here

Install SurrealDB, run it locally or in the cloud, and write your first query.

I want to build something

Connect an SDK, model your data, secure it, and put it into production.

I need the details

Every SurrealQL statement, protocol message and SDK method.

Your first query

SurrealQL is SQL-shaped, so most of it reads as you would expect. What it adds is the ability to relate records directly and traverse those relations in the same statement, without a join table or a second database. Edit and run this:

The querying guide covers the language properly, and data models covers what else the engine stores.