Weaviate is an open-source vector database designed for AI-native applications, built around the idea that vectors are first-class citizens of the data model. It combines vector storage and search with a flexible schema, native hybrid search, and built-in vectorizer modules that can generate embeddings automatically on ingestion.
A distinguishing feature is its modular, batteries-included approach. Rather than requiring a separate embedding pipeline, Weaviate can integrate with embedding models directly, so you can send raw data and let the database handle vectorization. It offers both GraphQL and REST APIs, native hybrid search that fuses vector and keyword results, and strong metadata filtering treated as part of the core query engine rather than an add-on.
Weaviate is often cited for capabilities like native multi-tenancy with per-tenant shards, tenant lifecycle management including offloading inactive tenants to cold storage, and a knowledge-graph-like schema that links objects through references. Available both as open-source software for self-hosting and as a managed cloud service, it is a common choice for teams building semantic search, retrieval-augmented generation, and agentic applications that want retrieval logic close to the data.