SurrealDB vs.
vector databases

Q: Can I migrate from a vector database to SurrealDB?

Yes. SurrealDB supports the same embedding formats and similarity functions, so existing vectors import directly and gain graph traversal, structured queries and ACID transactions.

Q: Do I still need a vector database if I use SurrealDB?

SurrealDB provides native vector indexing with HNSW and brute-force search, with vectors stored beside their source entities under full ACID guarantees, so it stands in for a separate vector store.

Start free with SurrealDB Explore the database

Vector databases. Pinecone, Chroma, Weaviate and Qdrant store embeddings, leaving the entities around them elsewhere.

SurrealDB. Native HNSW and brute-force vector search beside documents, graphs and structured data.

One database. Every model in one engine and one transaction, so a vector and its source document update together.

01 |THE LIMITATION

Vectors alone are not enough. Context takes more than similarity search.

Embeddings alone

Vector databases store embeddings, and leave the entities, relationships and metadata that give them meaning elsewhere.

Relationships out of reach

A user and their purchases live in separate systems, with no traversal between them.

Two systems, two commits

Updating a vector and its source document spans two stores, so one write can succeed while the other fails.

Present tense only

Vector databases keep the current embedding, with no history of how the knowledge changed.

02 |THE DIFFERENCE

Vectors as part of the whole. Embeddings beside the records they describe. One record, one transaction.

VECTORS

Co-located data

Embeddings beside their source documents, entities and relationships. One record, one transaction.

Explore vector search

QUERIES

Hybrid queries

Vector similarity, graph traversal, full-text search and structured filters in one SurrealQL statement.

Explore SurrealQL

TRANSACTIONS

ACID

Read-think-write loops commit atomically: memory and state in one transaction.

How transactions work

ACCESS

Unified permissions

One permission model for documents, graphs, vectors and memory, with RBAC and record-level access.

Explore permissions

03 |HOW IT WORKS

One query, every signal. Vector similarity, a graph hop and a structured filter in one SurrealQL statement, on one index set, in one transaction.

1-- Similar documents by one team, this quarter:
2-- vector, graph and filter in one pass
3SELECT
4    title,
5    vector::similarity::cosine(embedding, $q) AS score,
6    <-authored<-person.name AS authors
7FROM document
8WHERE embedding <|10|> $q
9    AND <-authored<-person->member_of->team CONTAINS $team
10    AND published > time::now() - 90d
11ORDER BY score DESC;

From vector search to structured memory

Knowledge graph

Entities, relationships and facts, the structural context similarity search alone cannot give.

Hybrid retrieval

Temporal awareness

Three clocks record when a fact held, when it was learnt and when the system stored it, and a new value closes the old one rather than overwriting it.

TRUSTED BY

Enterprise teams building on SurrealDB

See more 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

Vector databases,answered

How does SurrealDB's vector search compare to Pinecone or Weaviate?

SurrealDB runs native vector search beside documents, graphs, full-text and time-series in one engine. Pinecone and Weaviate are dedicated vector databases that lean on other systems for the rest, and on a sync layer between them.

Can I migrate from a vector database to SurrealDB?

Do I still need a vector database if I use SurrealDB?

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

Dedicated clusters by region, or serverless on SurrealDB Earth.

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

The multi-model database for AI. Documents, graphs, vectors and time-series in one engine, one query, one transaction.