Qdrant

Qdrant is a vector similarity search engine and vector database, written in Rust, that stores and searches "points" - vectors plus an arbitrary payload - with an emphasis on rich, extended filtering alongside the similarity search itself (combining semantic matching with structured metadata filters).

It runs as a server (via Docker or other deployment) with official client libraries for Python, JavaScript/TypeScript, Go, Rust, .NET/C#, and Java, or as Qdrant Edge, a lightweight embedded variant that runs inside your application process for edge or resource-constrained environments and can sync with a full Qdrant server. Qdrant also publishes ready-made "agent skills" that bring its vector-search tooling directly into AI coding assistants.

It's available self-hosted or as a managed Qdrant Cloud service with a free tier. Qdrant fits teams building semantic search, recommendation, or other embedding-based matching features that also need filtering on structured metadata alongside the vector search itself.