Weaviate is an open-source vector database known for its first-class hybrid search (combining dense vector similarity with sparse keyword matching via BM25) and its built-in module system that handles embedding generation for you. Unlike Pinecone which is pure vector-search-as-a-service, Weaviate is a full platform you can run locally or in their managed cloud.
Weaviate defines a schema with classes and properties. You can either push pre-computed embeddings or let Weaviate call an embedding model (OpenAI, Cohere, local transformers) automatically via its text2vec modules. Queries support hybrid mode (alpha parameter controls vector-vs-keyword weight), metadata filters, and aggregation. GraphQL API is the primary query interface, though REST is also available.
Open source: free, self-host on your own infrastructure. Weaviate Cloud: Starter $25/month, Standard $295/month, Enterprise custom. Self-hosting costs depend on your infrastructure — a single VM can handle millions of vectors.
Instabase, Morningstar, Chargebee, Redis, and many mid-sized enterprises that need hybrid search. Popular in the RAG community for its built-in embedding modules.