SurrealDB vs.
traditional databases

Q: What about bi-temporal versioning?

SurrealDB records transaction time with VERSION on a versioned store; valid time is modelled with valid_from and valid_until fields, and Agent Memory tracks both for facts.

Q: How does Gen 3 storage reduce costs?

Commodity object storage in place of provisioned disk, stateless compute that scales to zero when idle, and data routed through object storage rather than across zones. Idle development and staging environments cost nothing.

Start free with SurrealDB Explore the database

Generation one. Single-model databases, one per workload.

Generation two. Specialist stores stitched together in application code.

Generation three. Every model in one engine.

01 |THE LIMITATION

One model each. Graphs by workaround, vectors by extension, the present tense only.

Graphs by workaround

Postgres reaches a traversal through joins and CTEs rather than a graph model.

Vectors as an extension

pgvector bolts vectors onto a relational core that was not designed for hybrid or agent queries.

Current state only

Traditional databases hold the present value, so the question of what was known when has no answer.

02 |THE DIFFERENCE

Everything in one engine. Documents, graphs, vectors, time-series and geospatial, in one language and one transaction.

MODELS

Native multi-model

Documents, graphs, vectors, time-series and geospatial, all first-class in one engine.

Explore the data models

QUERIES

SurrealQL

One SQL-like language: joins, traversals and vector search share a syntax.

Explore SurrealQL

TRANSACTIONS

ACID across models

One transaction boundary. A document and its graph relationships update atomically.

How transactions work

AGENTS

Agent-ready

Agent Memory: knowledge graphs, entity extraction and temporal facts, in the database.

Explore Agent Memory

03 |HOW IT WORKS

One statement, three models. A customer record, the graph of what they bought and the vector that describes them, read together in one SurrealQL statement.

1-- One statement across three models: the record, its graph, its vector
2SELECT
3    name,
4    ->purchased->product.name AS bought,
5    vector::similarity::cosine(embedding, $q) AS fit
6FROM customer
7WHERE embedding <|5|> $q
8    AND ->purchased->product->belongs_to->category CONTAINS $category
9ORDER BY fit DESC;

Three generations of database infrastructure

Gen 1: Monolith

Compute and storage coupled in one machine; scaling means a bigger box. MySQL, self-hosted PostgreSQL, classic Oracle.

Gen 2: Proprietary loose coupling

Cloud vendors separated storage from compute, in proprietary formats with cloud lock-in. Amazon Aurora, Oracle Exadata, Google AlloyDB.

Gen 3: Open storage on the lake

Data in commodity object storage in open formats, with stateless elastic compute over it. Databricks Lakebase, SurrealDB.

TRUSTED BY

Enterprise teams building on SurrealDB

See all case studies

Three data stores collapsed into one

An AI assistant for 10,000 technicians

Nine backend tools consolidated into one

Low-latency RAG, under the customer's control

45 million recommendation queries a month

Traditional databases,answered

Can SurrealDB replace Postgres or MySQL?

Yes. Relational queries, SQL-like syntax, ACID transactions and structured schemas, plus native graph traversal, vector search, temporal queries and agent memory.

What about bi-temporal versioning?

How does Gen 3 storage reduce costs?

THE PLATFORM

Everything an application and its agents know. Five surfaces, one engine.

Database

Document, graph, vector, time-series and relational in one engine.

Five data models as dotted tiles: documents, graph, vector, time-series and relational

Agent Memory

What an agent learns, with its source and its time, in the same engine.

A timeline of remembered facts, each with its source

Cloud

Managed clusters in the regions you choose, scaled on demand.

Clusters in three regions on a world map, each running or scaling

Studio

Query, explore and design the schema from the browser.

MCP

Every model that speaks MCP reaches the database and the memory directly.

Three models connected through MCP to the database and Agent Memory

IN PRODUCTION

Trusted at scale. Samsung, Nvidia, Verizon, Tencent and Walmart run on SurrealDB.

14,000+

Developers building on SurrealDB Cloud

4M+

Developers building on SurrealDB worldwide

FROM THE TEAMS

SurrealDB gives us a foundation where we can unify semantic search, knowledge graphs, and AI-driven decision making without stitching together multiple systems. Collapsing responsibility into SurrealDB has become our default engineering posture.

Justin Foley

VP of Engineering, Later

GET STARTED

Move to Generation 3. ACID rigour with the elasticity, durability and economics of the cloud.