# lancedb/lancedb

Developer-friendly OSS embedded retrieval library for multimodal AI. Search More; Manage Less.

Repository: https://github.com/lancedb/lancedb
Canonical: https://ross.abutalabs.com/products/lancedb
Homepage: https://lancedb.com/docs
Language: Rust
License: Apache-2.0
License Family: permissive
Topics: approximate-nearest-neighbor-search, image-search, nearest-neighbor-search, recommender-system, search-engine, semantic-search, similarity-search, vector-database
Last push: 2026-08-26T23:48:01+00:00

## Health v2 (maintenance only)
Score: 92/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 99, release rhythm 85, longevity 91
- inputs: {"age_days": 1283, "days_push": 7, "days_rel": 23, "gap_med": 0.0, "n_releases_24m": 341}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 11284, forks 1027 (observed 2026-08-28T04:10:46.652245+00:00)

## What it is
LanceDB is an open-source embedded vector database and multimodal AI lakehouse built on the Lance columnar format, written in Rust with Python, TypeScript, and Rust SDKs. It supports fast vector similarity search, full-text search, and SQL over billions of vectors and multimodal data without managing separate server infrastructure.

## Use cases
- store and search embeddings for a RAG application
- build semantic search over documents, images, or videos
- run a local vector database without a separate server
- find similar images by embedding similarity
- build a recommendation system with nearest-neighbor search
- curate and version training data for ML pipelines
- combine vector search with SQL filtering and full-text search

## When to choose
- you want an embedded, serverless vector store that runs locally or on object storage
- you need multimodal data (text, images, video, point clouds) alongside vectors in one table
- you want zero-copy versioning of datasets without extra infrastructure
- you need fast ANN search at scale with GPU index building

## When to avoid
- you need a fully managed multi-tenant database service with built-in access control (consider LanceDB Enterprise or a hosted DB)
- your workload is purely relational OLTP with no vector or search needs
- you require rich secondary-index transactional semantics of a traditional RDBMS

## Facets
- artifact type: library
- maturity: active
- function: vector-database, search-engine, database, rag
- domain: databases, machine-learning, artificial-intelligence, data-science
- platform: rust, python, cross-platform, cli
- tags: embedded-database, similarity-search, multimodal-ai, lance-format, ann-search, full-text-search, serverless, search, retrieval-augmented-generation, nodejs

## Member repositories
- lancedb/lancedb (main) score 92

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:46.652245+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:15:59.270083+00:00, confidence not recorded.
  - readme: https://github.com/lancedb/lancedb (fetched 2026-08-28T04:10:46.652245+00:00, sha 5ecb30ea9e80)
  - homepage: https://lancedb.com/docs (fetched 2026-08-29T08:14:46.868068+00:00, sha 24a26d26940c)
  - site_page: https://docs.lancedb.com/quickstart (fetched 2026-08-29T08:14:46.877685+00:00, sha 4405304875eb)
  - site_page: https://docs.lancedb.com/integrations (fetched 2026-08-29T08:14:46.879787+00:00, sha a3326e49ee46)
  - site_page: https://docs.lancedb.com/integrations/lerobotdataset (fetched 2026-08-29T08:14:46.881540+00:00, sha 88f034a404d6)
  - site_page: https://docs.lancedb.com/geneva/getting-started (fetched 2026-08-29T08:14:46.884192+00:00, sha 5437c2a01719)
  - registry_crates: https://crates.io/api/v1/crates/lancedb (fetched 2026-08-29T08:14:46.886114+00:00, sha 06cb5143fc68)
- Data as of 2026-08-30T08:39:29.467469+00:00.