qdrant — Qdrant is a high-performance vector similarity search engine and database written in Rust, designed for AI applications requiring semantic search and large-scale vector management.

Qdrant is a high-performance vector similarity search engine and database written in Rust, designed for AI applications requiring semantic search and large-scale vector management.

About this project

Qdrant is a production-ready vector database and search engine that allows users to store, search, and manage vectors accompanied by JSON payloads. Written in Rust, it is designed for high load and reliability. Key capabilities include: - Search Types: Supports dense vectors for semantic similarity, sparse vectors for full-text search, and multi-vector search (e.g., for ColBERT). - Hybrid Search: Combines multiple vector types using fusion strategies like Reciprocal Rank Fusion (RRF) and Distribution-Based Score Fusion (DBSF). - Advanced Filtering: Provides rich filtering on payloads using keyword matching, numeric ranges, and geo-locations. - Performance Optimization: Includes built-in vector quantization to reduce RAM usage, SIMD hardware acceleration, and GPU support for indexing. - Deployment Options: Offers a client-server architecture, a lightweight "Qdrant Edge" for resource-constrained environments, and a managed cloud service. - Scalability: Supports horizontal scaling via sharding and replication with zero-downtime updates. - Interfaces: Provides both REST (OpenAPI 3.0) and gRPC APIs, with official client libraries for Python, Rust, Go, JavaScript/TypeScript, .NET/C#, and Java. - Additional Features: Includes a Web UI for data interaction, multitenancy support, and observability tools for monitoring.