About SurrealDB
SurrealDB is a multi-model database that stores documents, graphs, vectors and relations in one engine. It can run on your own servers, inside an app, or as a managed cloud service.
What it does
SurrealDB combines several data models in one database, so an app does not need separate stores for documents, graph links, vector search and time-series data. Its transactions are fully ACID. A separate agent memory layer connects and retrieves context from wherever your data lives, and each fact keeps its source.
Key features
- Documents, graphs, vectors and time-series in one engine
- Live queries that update as data changes
- Fully transactional, ACID-compliant data
- Serverless option that scales to zero when idle
- Agent memory that keeps the source of each fact
Who it's for
Developers building AI agents or AI-heavy apps that need one place for structured data, vector search and memory. It also suits teams that want a single database engine able to run on their own servers or in the cloud.
Pricing and good to know
Self-hosted, embedded and managed cloud options are all offered. The serverless option, SurrealDB Earth, runs on Cloudflare and starts within about 300 ms on the first query, according to the site. Plan details are on the pricing page.
Highlights
- Documents, graphs, vectors and time-series in one engine
- Live queries that update as data changes
- Fully transactional, ACID-compliant data
- Serverless option that scales to zero when idle
- Agent memory that keeps the source of each fact
