Weaviate

Weaviate is an open-source vector database that combines vector search with structured filtering and a GraphQL API. It can be self-hosted or used as a managed cloud service, offering flexibility for different deployment requirements.

Distinctive features include: built-in vectorization modules (can generate embeddings automatically), GraphQL-based query interface, hybrid search combining BM25 and vector similarity, multi-tenancy support, and generative search (combining retrieval with LLM generation). Weaviate also supports various distance metrics and index types.

Weaviate is popular among teams wanting more control over their vector infrastructure or needing self-hosted options. The built-in vectorizers simplify pipelines by eliminating separate embedding generation steps. Engineers should evaluate: deployment preferences (cloud vs self-hosted), query patterns, scaling requirements, and whether built-in features like generative search fit their use case.