The unified data
layer for AI

Graphs, vectors, documents, and relational data in one engine.
Agent memory that connects and retrieves context wherever your data lives.

Explore the database Explore agent memory Get started

SOC 2 Type 2

GDPR

Cyber Essentials Plus

ISO 27001

CHOOSE YOUR PATH

What are you building?

One platform, two ways in: a database you model yourself, or a memory layer that does the AI modelling for you.

Database

One engine for your data - modelled by you, queried your way.

Documents, graphs, vectors, and SQL in one ACID engine

You design the schema and own the model

In production at Samsung, Verizon, and Tencent

Explore the database

The database: documents, graphs, vectors, and time-series in one engine

Agent memory

Memory for your agents - built for you from your own data.

Connect and retrieve context in one line of code

Assembles itself from conversations, documents, and your systems

In early access - join now

Explore agent memory

Agent memory: persistent agent memory built on the database

THE PROBLEM → THE FIX

Agents fail due to context. Not models.

Your model can reason. It just has nothing reliable to reason over. Each failure below has a structural fix - built into the engine.

Context that leaks at every seam

Data flows between six systems. Relationships, history, and metadata fragment at every boundary, silently eroding accuracy.

App state

User session

Vector DB

Embeddings

Graph DB

Relations

Doc store

Documents

Auth

Identity

Data fragments at every system boundary

Token bills that balloon

Every store returns its own copy of the same context. Duplicate chunks inflate each prompt - and the token bill with them.

Vector DB

same fact

Graph DB

same fact

Doc store

same fact

Prompt

One fact, 3x the tokens

Latency that compounds

Every system adds a network hop. Round trips stack under load until agents miss their response window.

Agent

Vector DB

Graph DB

Doc store

Auth

Total

Cumulative latency

Six systems to keep alive

Six failure modes, six monitoring setups, six sets of credentials. The glue code becomes the product.

Vector DB

Similarity search

auth

config

monitor

Graph DB

Relationship store

auth

config

monitor

Doc store

Raw document index

auth

config

monitor

Auth service

Identity & tokens

auth

config

monitor

Cache layer

Hot key store

auth

config

monitor

Five systems. Five auth configs. Five things to scale, secure, and keep alive.

HOW IT FITS TOGETHER

One platform.
The database at the core.
Memory on top.

They're not two stacks to run: Agent memory, the agent memory layer, sits on top of SurrealDB, the database - one transaction boundary, one permission model, one deployment.

Applications

Agent memory

Entity extraction

Knowledge graph

Temporal facts

Hybrid retrieval

Database

Documents

Graphs

Vectors

Time-series

Auth

APIs

Distributed write nodes

Node A

Node B

Node C

Object storage (S3 / GCS / Azure Blob)

USE CASES

What teams build on one engine

Memory for agents and a multi-model database in one platform - each use case runs on the product built for it.

GraphRAG

Database

Retrieval that combines graph traversal with vector search for grounded, connected context.

Traverse relationships instead of returning disconnected chunks

Rank by semantic relevance and full-text in the same query

Lower latency, with no vector database stitched to a graph database

Explore the database

Agent memory

Persistent, queryable memory for agents - relationships, semantics, and recency in one store.

Knowledge-graph memory of entities, threads, and episodic timelines

Vector recall of past facts and decisions, per user or per tenant

Build memory yourself, or build your memory product on top of it

Explore Agent memory

Knowledge graphs

Database

Model entities and relationships natively - query structure and meaning together.

First-class nodes, edges, and properties

Semantic and full-text search over all your knowledge

Type-safe traversals in Rust, Go, TypeScript and Python

Explore the database

Company brain

Agent memory

One queryable source of truth across every document, person, tool, and decision.

Connect org structure, docs, and projects as a graph

Temporal awareness of when facts changed

Agents traverse sub-graphs and tenants with permissions intact

Explore Agent memory

PROVEN IN PRODUCTION

In production at
Samsung, Verizon,
Tencent & PolyAI.

Proof from production, every quote and figure tagged by product. The database runs in production across industries.

TRUSTED BY THE BEST

The SurrealDB integration was seamless and delivered performance on par with our internal stack. It proves that enterprises can bring their own knowledge base without sacrificing speed, quality, or control.

DATABASE

Colman Yau

VP of Engineering, PolyAI

SurrealDB is enabling the next phase of our product. It gives us the flexibility and graph-native capabilities we need to keep innovating for customers at every size and industry.

DATABASE

Or Weis

CEO & Co-founder, Permit.io

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

DATABASE

Lucy Egan

Aspire

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!

DATABASE

Sigismond

Community developer

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.

DATABASE

Trevor Parscal

Community developer

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.

DATABASE

Albert Marashi

Community developer

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

DATABASE

@BeniaminDudek

Community developer

PolyAI case study Tencent case study Later case study See all case studies

BY THE NUMBERS

700K

DATABASE

Users on Aspire

Scaled in 8 hours after replacing 5 backend tools with SurrealDB.

50M

DATABASE

Graph edges at Tencent

Across 8M nodes at 10,000+ QPS on a single context graph.

~30ms

DATABASE

RAG latency at PolyAI

Vector search on par with their internal stack - no perceptible delay.

BENCHMARKED IN THE OPEN

Faster with every release

Performance measured in the open. SurrealDB 3.x against 2.x with the open-source crud-bench harness on the same hardware.

View the benchmarks

↑ 31%

DATABASE

Faster CRUD

Mean across creates, reads, updates, deletes - 2.x to 3.x

↑ 58%

DATABASE

Faster batches

Mean across batched operations - 2.x to 3.x

↑ 11894%

DATABASE

Faster full-table scans

Mean across non-indexed read scans - 2.x to 3.x

↑ 136%

DATABASE

Faster indexed queries

Mean across indexed read scans - 2.x to 3.x

CASE STUDIES

Saks Fifth Avenue uses SurrealDB's vector search and graph capabilities to deliver real-time, AI-powered personalisation across 5 million luxury customers and 45 million monthly product-recommendation queries.

Read case study

THE COMMUNITY