Description

LanceDB is an open-source vector database developed for AI applications that store, manage, query, and retrieve embeddings on large-scale multi-modal data. The core of LanceDB is written in Rust and built upon the Lance format, which offers performant ML workloads and fast random access. This database is designed to be easy-to-use, scalable, and cost-effective for both storage and compute. Available in two flavors - OSS (open-source) and Cloud (SaaS) - LanceDB supports multi-modal data storage alongside embeddings and metadata, providing automatic data versioning and blazing fast retrievals and filtering via LanceDB. The OSS version is embedded with in-process hosting, while the Cloud version runs serverless in the cloud for scalability without breaking the bank.