LanceDB is an open-source database designed for vector-search with persistent storage, making the retrieval, filtering, and management of embeddings easier. Key features include production-scale vector search without server management, support for multiple data types, vector similarity search, full-text search, SQL, and native compatibility with Python and Javascript/Typescript. It offers zero-copy, automatic versioning, and GPU support for building vector indexes, along with integrations with LangChain, LlamaIndex, Apache-Arrow, Pandas, Polars, DuckDB, and others.

2m read timeFrom github.com
Post cover image
Table of contents
Quick StartBlogs, Tutorials & Videos

Sort: