Qdrant

Health v2 · maintenance only

Adoption not part of the score

34210 stars · 2615 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

Qdrant is an open-source, production-ready vector database and similarity search engine written in Rust, supporting dense, sparse, and multivector embeddings with rich JSON payload filtering and hybrid search. It ships as a self-hostable service, a managed cloud offering, client libraries, and an MCP server for AI agent integration.

Use cases

• build semantic search over documents with embeddings
• store and query vectors for a RAG pipeline
• recommend similar products with filtered similarity search
• hybrid search combining dense and sparse embeddings with reranking
• find nearest neighbors in image or text embedding spaces
• self-host a vector database for AI applications
• give AI agents vector search via MCP

When to choose

• you need production-grade vector search with metadata filtering at scale
• you want hybrid search with dense, sparse, and late-interaction (ColBERT) vectors
• you need a self-hosted or managed vector store with client libraries in many languages
• you need filterable ANN search without recall loss

When to avoid

• you only need exact k-NN on small datasets where a library like FAISS suffices
• your workload is purely relational/SQL with no vector similarity needs
• you need a full general-purpose document search engine with text analysis rather than vector retrieval

Facets

service · maturity stable

vector-database search-engine database rag databases artificial-intelligence machine-learning rust self-hosted cloud python cross-platform hnsw embeddings similarity-search hybrid-search semantic-search recommender-systems mcp-server ann search retrieval-augmented-generation docker

Member repositories

For agents

markdown · JSON · MCP: product_card(name="qdrant/qdrant")