qdrant and qdrant-client
The database engine and its official Python client library are complements designed to be used together, where the client provides programmatic access to Qdrant's vector search functionality.
About qdrant
qdrant/qdrant
Qdrant - High-performance, massive-scale Vector Database and Vector Search Engine for the next generation of AI. Also available in the cloud https://cloud.qdrant.io/
Built in Rust for reliability under high load, Qdrant supports extended JSON payload filtering alongside vector similarity search, enabling semantic matching with rich metadata constraints. It offers both REST and gRPC interfaces, with official clients for Python, Go, Rust, JavaScript, Java, and .NET, plus integrations for semantic search, image retrieval, recommendations, and anomaly detection use cases.
About qdrant-client
qdrant/qdrant-client
Python client for Qdrant vector search engine
Provides type-safe bindings for all Qdrant API methods with dual REST and gRPC transports, plus a local in-memory or disk-persisted mode for development without a server. Built-in embedding inference via FastEmbed (CPU/GPU) or Qdrant Cloud models enables end-to-end vector workflows in a single client, simplifying document upload and semantic search operations. Supports both synchronous and asynchronous request patterns with helper methods like `upload_collection` that handle chunking and batch operations automatically.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work