Qdrant

开源的向量数据库和向量相似性AI搜索引擎

官网公开介绍

Qdrant is an Open-Source Vector Search Engine written in Rust. It provides fast and scalable vector similarity search service with convenient API.

High-Performance Vector Search at Scale

Qdrant helps you build the AI retrieval you want. Ship high performance, full-feature vector search at any scale and with any deployment model.

Build for Production-Grade AI Search

Engineered for real-time retrieval with the speed, accuracy, and scale that modern AI demands.

Expansive Metadata Filters

Store metadata in JSON and use advanced filters, such as nested, text, geo, has_vector, and more.

Native Hybrid Search (Dense + Sparse)

Blend keyword and vector search in one query – use dense or sparse vectors. Supports BM25, SPLADE++, and miniCOIL.

Built-in Multivector

Set new standards for relevance; make the retrieval layer more expressive, flexible, and multimodal with multiple vectors per object.

Efficient, One-Stage Filtering

Filters are applied during HNSW traversal — no pre- or post-filtering. High recall with low latency, even under complex conditions.

Full-Spectrum Reranking

Infuse business logic with score boosting, achieve token-level precision with late interaction models (e.g. ColBERT), diversify results with Maximum Marginal Relevance (MMR)

One Engine, Endless Applications

Powers AI Trip Planner on billions of reviews and images, driving 2-3x revenue.

Powers Breeze AI with real-time, personalized responses and deep contextual awareness.