Weaviate is a vector database made by Weaviate BV. A vector database stores each item together with an embedding, a list of numbers produced by an AI model that represents the item's meaning, and finds the stored items whose vectors are closest to the vector of a query. This is the basis of semantic search and of RAG (retrieval-augmented generation, where a chatbot first fetches relevant text and then answers from it).
Weaviate stores both objects and their vectors, so vector search can be combined with structured filters, and it offers hybrid search. Built-in embedding services turn text and images into vectors without a separate pipeline. The Query Agent turns natural-language questions into database queries. Multi-tenancy keeps many customers' data in separate tenants inside one cluster.
The database code is open source, mostly under the BSD-3-Clause license, with some enterprise code under a separate license; it can run on a company's own servers. Weaviate Cloud is the hosted version, with a free plan, a usage-billed monthly Flex plan and prepaid Premium contracts.