The Embedded Vector Database

LanceDB is an open source embedded vector database. Run it inside your application or notebook. No clusters, no containers, no separate service. Install the library, point it at a local path, start building.

Why teams switch

Open source, vendor-flexible

Apache 2.0 license. Audit the code. Run on your hardware. Built on the open Lance format with Arrow types.

Embedded simplicity

Runs in-process. No Docker, no K8s, no separate service. Native Python and TypeScript clients.

Same code, local to cloud

The embedded library and managed service share the same format and query engine. Swap endpoints, not code.

Full-text + hybrid search, native

Vector, full-text, and SQL queries in one system. Built in, not bolted on.

Comparison

	Other Embedded Options	LanceDB
Cost	Varies. Some require managed service for production.	Object storage pricing. OSS is free.
Scale	Often limited by single-machine memory.	20 PB largest table. Same format from laptop to petabyte.
Search	Typically vector-only	Native vector, full-text, and SQL hybrid in one query.
Data model	Embeddings only. Raw data elsewhere.	Raw data (blobs), embeddings, and features in one table.
Purpose	Local development, prototyping.	Local to production on the same engine.
Best for	Prototyping only.	Embedded development that scales to production.

The Power of the Lance Format

Vector Search

Fast scans and random access from the same table — no tradeoff
Zero-copy access for high throughput without serialization overhead

Multi-Modal

Raw data, embeddings, and metadata in one table — not pointers to blob storage
No separate metadata store to keep in sync

Enterprise-Grade Requirements

Security

Granular RBAC, SSO integration, and VPC deployment options.

Governance

Data versioning and time-travel capabilities for auditability.

Support

Dedicated technical account management and guaranteed SLAs.