lancedb/docs/web/index.mdx

---
title: LanceDB
sidebarTitle: "LanceDB"
description: "Multimodal lakehouse for AI."
icon: "/static/assets/logo/lancedb-icon-gray.svg"
keywords: ["multimodal lakehouse", "training", "feature engineering", "search", "open source", "oss"]
---

**LanceDB** is a [multimodal lakehouse](https://lancedb.com/blog/multimodal-lakehouse/) for AI teams that need
one data layer for curation, feature engineering, search and retrieval, and model training.
It is built on top of [Lance](/basics/lance), an open-source lakehouse format designed for multimodal AI data.

Move from data exploration to model training on one unified platform without needing to manage a
fragmented stack of storage, feature, retrieval, and training systems.

## Train on the right data faster {#train-on-the-right-data-faster}

Search, curate, and train at petabyte scale, with all your multimodal data in one table.

Preparing training data and running experiments slow down when raw data, metadata, embeddings, features, and governance
artifacts live in separate systems. LanceDB keeps them together in one versioned multimodal table, so AI teams spend less
time stitching infrastructure together and more time improving datasets, testing features, and keeping GPUs fed.

## Accelerate the data flywheel for model research {#accelerate-the-data-flywheel-for-model-research}

Keep your AI data in one table, and give each agent-driven experiment its own isolated branch.
Find the failure, fix the data, and retrain.

<Steps>
  <Step title="Find a curated training subset for every failure case">
    Start from one failed grasp or evaluation example. [Search](/search/) multimodal data, metadata, and embeddings
    together, retrieve similar examples, and [curate](/curation) them into a targeted training subset.
  </Step>
  <Step title="Data sandboxes for safe agentic experimentation">
    Don't let an agent silently corrupt your data and ruin a training run. Give each agent its own
    [branch](/tables/branching), a writable fork of the table, to experiment on, and review the changes before
    they reach `main`. See [Run experiments on branches](/agent-branch-experiments).
  </Step>
  <Step title="A single dataset for every modality">
    Keep episodes, frames, metadata, features, and embeddings together in one schema-enforced table, without
    pipelines to keep them in sync. Table [versions](/management/table#versioning) let you trace each training
    run back to the exact data it used.
  </Step>
  <Step title="More hillclimbing, less babysitting preprocessing jobs">
    Define the [computed columns](/feature-engineering/add-computed-column) and
    [materialized views](/feature-engineering/materialized-views) your experiment needs, such as extracted features
    or embeddings, and refresh them as your data changes. On LanceDB Enterprise, column refreshes run as background
    jobs on the cluster instead of in your code.
  </Step>
  <Step title="Higher GPU utilization, faster convergence">
    Keep GPUs fed. [`StreamingDataset`](/training/data-loading) streams multimodal training data into PyTorch with
    globally shuffled, deterministic batches, combining fast random access with high-throughput scans.
  </Step>
</Steps>

Use the same table to curate training data, add derived features, retrieve examples, and feed training jobs that rely on expensive GPUs.
Training workloads can sample, shuffle, and scan projected columns from local storage or object storage, then assemble
GPU-ready batches from a tagged dataset version. Training pipelines need curation, derived features, reproducible
splits, fast random access, and a clean path into frameworks such as PyTorch -- LanceDB keeps all of it connected
through the same table model, whether a workflow uses one table or several related ones.

## Architecture {#architecture}

LanceDB Embedded and Enterprise are both built on [Lance](/basics/lance), which stores table data, versions, and
indexes in open, columnar files. The format and table model are the same across both offerings -- what changes is
how storage and compute are organized around them.

![LanceDB suite: LanceDB Embedded for search, and LanceDB Enterprise for search, data exploration and curation, feature engineering, and training, both on the Lance format](/static/assets/images/overview/lancedb-suite.svg)

Embedded reads and writes tables on local disk or in object storage from inside your application process.
Enterprise keeps durable table data and index artifacts in object storage, independently of the machines serving
requests, and separates query serving, read execution and caching, and indexing and background work into their
own independently sized components. Caches accelerate reads; they are never the durable source of truth. With
multiple serving replicas, Enterprise routes traffic to surviving nodes when one fails; losing a cache node
temporarily raises latency while its replacement warms, rather than causing an outage.

Sharing a format does not mean an Enterprise deployment automatically opens an existing Embedded database --
plan how to move or connect your data and configure access for your deployment.

## Embedded and Enterprise {#embedded-and-enterprise}

LanceDB ships as two offerings on the same Lance format and table model. Start locally with Embedded, then move to
Enterprise when your team needs distributed scale, managed infrastructure, private deployment, or higher-throughput
curation, feature engineering, search and retrieval, and training workflows.

### 1\. LanceDB Embedded {#1-lancedb-oss}

**LanceDB Embedded** is an open-source, embedded retrieval library, with client SDKs in Python, TypeScript and
Rust -- the application itself plans queries, builds indexes, and runs background work, with no separate database
server to deploy. This fits local retrieval applications, notebooks, development, and single-node data pipelines,
where your application controls its own compute resources and [table maintenance](/maintenance/compaction).

![LanceDB Embedded: the LanceDB SDK and the Lance format run in your application, reading and writing tables on local disk or in object storage.](/static/assets/images/overview/lancedb-embedded.png)

### 2\. LanceDB Enterprise {#2-lancedb-enterprise}

[LanceDB Enterprise](/enterprise) is a petabyte-scale (and beyond), distributed **multimodal lakehouse** platform
built for search, curation, feature engineering, and high-throughput training data access workflows on top of the
same core table abstraction, without building bespoke infrastructure. Applications connect through a LanceDB SDK
while the service handles requests across a fleet, with query serving, read execution and caching, and indexing
and background work sized and scaled as separate components. A control plane governs configuration, identity
and cluster lifecycle, separately from the data-plane components in the diagram below. See
[Distributed search](/enterprise/distributed-search) for how query nodes and plan executors divide a request.

![The LanceDB SDK connects to LanceDB Enterprise, where a query fleet, executor and cache fleets, the realtime engine, and a jobs fleet run on top of the Lance format and object storage.](/static/assets/images/overview/lancedb-distributed.png)

Teams choose Enterprise to:

- **Serve growing workloads** -- add query-serving capacity and let Enterprise cache frequently read table data near
  compute as traffic grows.
- **Reduce operational work** -- let the service handle background [compaction](/maintenance/compaction) and
  [cleanup](/maintenance/cleanup) instead of scheduling them in application code.
- **Fit deployment requirements** -- run a managed deployment or bring LanceDB into your own cloud account, with
  OpenTelemetry metrics and traces.
- **Connect data workflows** -- use the same tables for retrieval, curation, and training data access, with
  [SQL](/basics/sql), [feature engineering](/feature-engineering), and [Geneva](/geneva) feature backfills.

To set up LanceDB Enterprise in your organization, [reach out to us](https://www.lancedb.com/contact).

### Compare features {#compare}

| Area | Feature | Embedded | Enterprise |
| --- | --- | :-: | :-: |
| **Search** | [Vector search](/search/vector-search) | ✅ | ✅ |
|  | [Full-text search](/search/full-text-search) | ✅ | ✅ |
|  | [Hybrid search](/search/hybrid-search) | ✅ | ✅ |
|  | [Filtering](/search/filtering) | ✅ | ✅ |
|  | [Reranking](/reranking) | ✅ | ✅ |
|  | [Query plans](/search/explain) | ✅ | ✅ |
|  | [Distributed search](/enterprise/distributed-search) | ❌ | ✅ |
| **Indexing** | [Vector indexes](/indexing/vector-index) | ✅ | ✅ |
|  | [Full-text indexes](/indexing/fts-index) | ✅ | ✅ |
|  | [Scalar indexes](/indexing/scalar-index) | ✅ | ✅ |
| **SQL and curation** | [SQL (Arrow Flight SQL)](/basics/sql) | ❌ | ✅ |
|  | [Aggregation and profiling](/curation/aggregate) | ❌ | ✅ |
|  | [Sampling](/curation/sampling) | ✅ | ✅ |
|  | [Joins](/curation/join) | ❌ | ✅ |
|  | [Ranking and deduplication](/curation/windowing) | ❌ | ✅ |
|  | [Near-duplicate detection](/curation/dedupe) | ❌ | ✅ |
|  | [Chunking](/curation/chunking) | ❌ | ✅ |
|  | [Changes since a version](/curation/changes) | ❌ | ✅ |
|  | [Views](/management/view) | ❌ | ✅ |
| **Feature engineering** | [Computed columns, with refresh](/feature-engineering/add-computed-column) | ✅ | ✅ |
|  | [Materialized views](/management/materialized-view) | ❌ | ✅ |
|  | [Embedding registry](/embedding) | ✅ | ✅ |
| **Training** | [PyTorch data loader](/training/data-loading) | ✅ | ✅ |
|  | [Permutations](/training/permutation) | ✅ | ✅ |
|  | [Distributed training](/training/distributed-training) | ✅ | ✅ |
| **Data management** | [Branches](/tables/branching) | ✅ | ✅ |
|  | [Compaction and cleanup](/maintenance/compaction) | ✅ | ✅ |
|  | [Background jobs (managed indexing and maintenance)](/maintenance/background-jobs) | ❌ | ✅ |
|  | [Real Time engine (streaming writes and reads)](/tables/realtime-engine) | ❌ | ✅ |
| **Security** | [Authentication](/enterprise/authentication) | ❌ | ✅ |
|  | [Authorization](/governance/authorization) | ❌ | ✅ |

## Projects using LanceDB {#projects-using-lancedb}

### Training {#training}

- [stable-worldmodel](https://github.com/galilai-group/stable-worldmodel) -- a platform for reproducible world-model
  research built on a LanceDB data layer, reporting 3-4x faster data loading on Push-T versus HDF5 / MP4 at a
  fraction of the disk footprint.
- [le-wm](https://github.com/lucas-maes/le-wm) -- a joint-embedding predictive world model from pixels, trained on
  the stable-worldmodel platform and its LanceDB data layer.
- [lerobot-lancedb](https://github.com/lancedb/lerobot-lancedb) -- a drop-in LanceDB backend for Hugging Face
  LeRobot datasets with faster loading across robotics datasets.

### Agentic applications {#agentic-applications}

- [OpenClaw LanceDB Plugin](https://docs.openclaw.ai/plugins/memory-lancedb) -- stores an agent's long-term memory
  in LanceDB, automatically recalling relevant memories before a model turn and capturing new facts afterward.
- [Cognee](https://github.com/topoteretes/cognee) -- an open-source AI memory platform that gives agents persistent
  memory across sessions, using LanceDB as one of its vector store backends.
- [CrewAI](https://github.com/crewAIInc/crewAI) -- a multi-agent framework that
  [rebuilt its agent memory on LanceDB](https://www.lancedb.com/blog/crewai-rebuilt-agent-memory-on-lancedb).

Across these projects, LanceDB and Lance provide the multimodal data layer that keeps raw observations, annotations,
features, and training or memory access patterns in one format instead of scattering them across task-specific
stores.

## Start with your workload {#start-with-your-workload}

<CardGroup cols={2}>
  <Card title="Ingest your data" icon="inbox" href="/ingestion">
    Create tables, append records, and update or delete rows as your data changes.
  </Card>
  <Card title="Build search and retrieval" icon="search" href="/search/">
    Use vector search, full-text search, hybrid search, reranking, filtering, and SQL.
  </Card>
  <Card title="Curate training data" icon="filter" href="/curation">
    Profile, sample, chunk, join, and deduplicate records before training on them.
  </Card>
  <Card title="Engineer features in place" icon="layer-group" href="/feature-engineering">
    Derive computed columns and materialized views without moving data elsewhere.
  </Card>
  <Card title="Load data into PyTorch" icon="boxes-stacked" href="/training/data-loading">
    Use LanceDB tables and permutations for projected, shuffled, random-access training reads.
  </Card>
  <Card title="Manage catalog objects" icon="folder-tree" href="/management">
    Create, list, and drop tables, namespaces, and views from Python, TypeScript, or SQL.
  </Card>
</CardGroup>