RAG e sistemi di conoscenza

LanceDB

lancedb/lancedb

An open-source, developer-friendly embedded retrieval library and multimodal AI lakehouse designed for fast, scalable, and production-ready vector search, built on the Lance columnar format.

★ 11,1KStelle
⑂ 994Fork
606Problemi aperti
RustLingua
Apache-2.0Licenza
Q@project.QualityScorePunteggio editoriale

Screenshot del progetto

Screenshot di LanceDB Screenshot di LanceDB Screenshot di LanceDB Screenshot di LanceDB Screenshot di LanceDB Screenshot di LanceDB

Panoramica

LanceDB is a multimodal data platform for AI/ML applications, built on top of the Lance columnar format. It enables developers to store, index, and search over petabytes of multimodal data and vectors. The platform supports vector similarity search, full-text search, and SQL queries. It offers automatic versioning, zero-copy data management, and GPU support for building vector indexes. LanceDB is available as a 100% open-source local library or as a managed cloud and enterprise solution.

Funzionalità principali

  • Fast vector search with state-of-the-art indexing
  • Comprehensive search supporting vector similarity, full-text search, and SQL
  • Multimodal support for text, images, videos, point clouds, and metadata
  • Zero-copy and automatic data versioning
  • GPU support for building vector indexes
  • Available as open-source local library or managed cloud service
  • SDKs for Python, TypeScript, Rust, and REST API

Requisiti, installazione e avvio rapido

Refer to the Quickstart documentation at https://docs.lancedb.com/quickstart to set up LanceDB locally. SDKs are available for Python, TypeScript, and Rust.

Utilizzo

Developers can interact with LanceDB via the Python SDK (https://lancedb.github.io/lancedb/python/python/), TypeScript SDK (https://lancedb.github.io/lancedb/js/globals/), Rust SDK (https://docs.rs/lancedb/latest/lancedb/index.html), or REST API (https://docs.lancedb.com/api-reference/rest). Detailed tutorials and recipes are available at https://github.com/lancedb/vectordb-recipes/tree/main.

Compatibilità del modello e casi d'uso

Not stated in the repository metadata.

Note su licenza e rischi

Licensed under Apache-2.0. License details available at https://api.github.com/licenses/apache-2.0.

Editorial verification 2026-08-09: repository URL, owner, description, license and repository statistics were reviewed. License metadata: Apache-2.0. README was fetched for the channel draft; re-check repository dependencies, releases and model terms before production use.

Rilascio e manutenzione

Not stated in the repository metadata.

EvalScope

modelscope/evalscope

★ 3,2KPython

Infino

infino-ai/infino

★ 67Rust