Overview
Weaviate is an open-source vector database with hybrid (vector + keyword) search, structured filtering, and optional vectorizer modules that embed your data for you.
It runs self-hosted with Docker or as managed Weaviate Cloud, and is a common retrieval layer for RAG and agents.
flowchart LR obj[objects + vectors] --> w[(Weaviate)] q[query vector] --> w w --> top[nearest objects]
The Code samples tab shows a bring-your-own-vectors flow.
When to use it
Choose Weaviate when you want hybrid search and rich filtering in one engine, with the option of built-in vectorizers — self-hosted or on Weaviate Cloud.