The multi-model database
for AI agents

Simplify your stack. Reduce complexity. Build faster.

Start for free

Learn more

The multi-model database
for AI agents

Simplify your stack. Reduce complexity. Build faster.

Start for free

Learn more

See why companies choose SurrealDB

Unify your data infrastructure
with a single platform

Unified multi-model database

Combines document, graph, time-series, relational, geospatial and key-value data models natively, without workarounds or added complexity.

AI native data layer

Purpose-built for AI and context-aware applications with integrated search and retrieval (vector, full-text, hybrid) that blend semantic, graph, and relational intelligence.

SELECT 
    supplier.{ name, address }, 
    vector::similarity::cosine(embedding, $ideal_spec) AS score
FROM manufacturer
WHERE geo::distance(location, $site) < 10000
ORDER BY score DESC
FETCH contacts;

Lower cost

Reduce total cost of ownership by eliminating the complexity of maintaining separate vector, graph, and document databases.

Unmatched scalability

Scale from a single node to a horizontally-scalable cluster with petabytes of data, without sharding.

Deploy anywhere

From in-memory and edge devices to self-hosted or global cloud infrastructure.

Real-time and event-driven

Built-in real-time subscriptions, event triggers, and streaming updates to power reactive, real-time experiences.

LIVE SELECT
  vector::similarity::cosine(details, $search)
  array::distinct(
    ->order->product,          -- products they bought
    <-order<-person,           -- other people who bought those
    ->order->product.{id,name} -- products those people bought
  ) AS recommend_products
FROM $user
WHERE details <|3|> $search;

Built-in security with RBAC, record-level permissions, fine-grained access controls, JWT authentication, multi-tenant isolation and built-in compliance (SOC 2, ISO 27001) keep data protected by default.

Deploy modern AI, at scale

Agentic and Gen AI

Build generative AI apps and multi-step agents on SurrealDB, unifying memory and action layers.

Learn more

GraphRAG

Supercharge retrieval-augmented generation with graph data models, for richer context and relevance.

Learn more

Knowledge Graphs

Combine vectors, events and user graphs for instant recommendations, personalisation and analytics.

Learn more

Too good to be true?

Don’t take our word for it.

Performance and benchmarks

As a multi-model database, SurrealDB is fast and scalable. See how it compares with other databases in our benchmarks.

Stars given to SurrealDB on GitHub

Join the community

Discover what our community has to say about their Surreal experience

Join our Discord

View us on GitHub

@UnchainedAlgo

I initially explored a more traditional stack: Neo4j for the graph database, RabbitMQ for a task queue, and Postgres with extensions for vector search. However, I realised SurrealDB could cover all of these needs, allowing me to consolidate the backend into a single dependency.

@ThatOneCalculator

I can never go back to Postgres lmao.

@Sigismond

SurrealDB fast tracked our progress with all its features. Graph links + Record links + Full text search + Vector embeddings and Vector search + Surreal WASM for offline-ability... just name it!!!! Thanks Surreal team for making it happen!!!!

Lucy Egan

Aspire

We replaced 5 backend tools with SurrealDB and scaled to 700,000 users in 8 hours.

@KenjiSuza

Fantastic! Graphrag, etc handles it all. Solid and fast

Chris Bruce

I know people love Supabase and Convex is blowing up. But I think SurrealDB is gonna be the real killer. More modern apps really should be using more graph databases as applications need to provide for more and more complex relationships.

@BeniaminDudek

So much easier to write advanced queries compared to SQL. Queries are actually readable! Community on Slack is also very helpful and includes SurrealDB developers.

@AjaySahoo

Surrealist’s way of exploring complex data feels simple as its intuitive, fast, and removes the need for complex queries.

@Michael Ncube

Great product, highly recommend

Trevor Parscal

I'm certainly enjoying that I can replace hundreds of lines of Postgres trigger and function code with a dozen characters in a DEFINE statement in SurrealDB.

@glitch.tar.lrz

The best next-gen multi-modal database that's out there.

Albert Marashi

SurrealDB has allowed our team to focus more on product and iteration without worrying about database constraints. We can do literally everything we need in SurrealDB.