SurrealDB is a multi-model database with document, graph, relational, and vector support. Free to self-host or use SurrealDB Cloud from $0.021/hr.
SurrealDB is a multi-model database designed to replace the common pattern of stitching together separate document, graph, relational, and vector databases within a single application. It provides document, graph, relational, and vector capabilities through one unified engine and query language called SurrealQL.
Founded in 2021 by brothers Tobie and Jaime Morgan Hitchcock and headquartered in London, SurrealDB has raised over 40 million dollars from investors including FirstMark and Georgian, and offers the database as an embeddable engine, a self-hosted server, or a managed cloud service called SurrealDB Cloud.
SurrealDB supports real-time live queries that push updates to connected clients as data changes, along with ACID-compliant transactions, built-in row-level authentication and permissions, and native vector search for AI and retrieval-augmented applications.
It can be embedded directly into an application as a single binary or WebAssembly module, deployed self-hosted, or run through SurrealDB Cloud, which offers vertical scaling on its Start tier and horizontal, multi-availability-zone scaling on its Scale tier. Some advanced cloud capabilities, such as database branching and forking, are planned for a future release.
SurrealDB can be self-hosted for free using its source-available core. SurrealDB Cloud offers a Start plan with one free instance and 1GB of free storage, then usage-based pricing beginning around $0.021 per hour.
The Scale plan is priced per node per hour, starting around $0.192, and adds multi-availability-zone deployment and multi-region disaster recovery. An Enterprise tier offers custom pricing for self-hosted or private-cloud deployments with SSO, audit logging, and SLA-backed uptime.
Yes, the core database can be self-hosted for free, and SurrealDB Cloud offers a free starting instance with usage-based pricing beyond that.
SurrealDB is used to store and query document, graph, relational, and vector data within a single database, commonly for serverless, JAMstack, and AI-powered applications.
SurrealDB was founded in 2021 by brothers Tobie Morgan Hitchcock and Jaime Morgan Hitchcock.
SurrealDB is headquartered in London, United Kingdom.
SurrealDB aims to cover relational and document use cases in one engine, though teams should evaluate it against their specific workload requirements before migrating.
Yes, SurrealDB includes native vector data types and search capabilities suited to AI and retrieval-augmented generation applications.
Yes, SurrealDB can be self-hosted on your own infrastructure at no licensing cost for the core engine.
SurrealQL is SurrealDB's purpose-built query language designed to express document, graph, relational, and vector queries.