Qdrant

Health v2 · maintenance only

Adoption not part of the score

34906 stars · 2721 forks observed · 2026-10-03

What it is AI-extracted, prompt v1, taxonomy v1, 2026-09-28, confidence 0.97

Qdrant is a high‑performance, open‑source vector similarity search engine and database written in Rust. It provides a production‑ready API for storing, searching, and filtering vectors with rich JSON payloads.

Use cases

• semantic similarity search over embeddings for document retrieval
• building recommendation systems based on vector similarity
• hybrid search that combines dense and sparse vectors with keyword filters
• filterable vector search with metadata payloads for e‑commerce or HR use cases
• RAG pipelines where a vector store retrieves relevant chunks before LLM generation
• offline, embedded search on edge devices using Qdrant Edge

When to choose

• You need fast, scalable ANN search with built-in payload filtering
• Your application requires hybrid queries or multivector support for late‑interaction models
• You want a self‑hosted or managed vector database with strong Rust performance guarantees
• You need a cloud‑native solution with easy Docker/Kubernetes deployment

When to avoid

• Your dataset is tiny and a simple relational DB suffices
• You only need full‑text search without vector semantics
• You have extremely low‑memory environments where even Rust overhead is too high
• You require a proprietary feature not yet supported by Qdrant (e.g., certain GPU‑accelerated indexes)

Facets

service · maturity active

vector-database search-engine rag artificial-intelligence machine-learning windows cloud self-hosted hnsw multivector payload-filtering hybrid-queries edge serverless enterprise hybrid-search filtering recommendation-system retrieval-augmented-generation data-engineering search linux macos docker

Member repositories

For agents

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