qdrant.io

Qdrant offers expansive metadata filters with JSON support, including nested, text, geo, and has_vector filters. It provides native hybrid search combining dense and sparse vectors, with support for BM25, SPLADE++, and miniCOIL. The engine includes built-in multivector capabilities for more expressive retrieval. It performs efficient one-stage filtering during HNSW traversal, ensuring high recall with low latency. Additionally, it supports full-spectrum reranking with score boosting, late interaction models like ColBERT, and Maximum Marginal Relevance (MMR).

Qdrant is used in various applications, including RAG and GenAI, AI agents, semantic search, recommendation systems, and data analysis. It powers AI trip planners on billions of reviews and images, real-time personalized responses for Breeze AI, and multi-agent platforms with real-time context across millions of conversations. It also supports Dust's AI agents platform with scalable vector search across thousands of data sources, and Lyzr's AI agents, reducing latency by 90% and increasing throughput by 150%.

The page does not provide specific pricing details. It mentions starting free in Qdrant Cloud and offers various deployment options including Qdrant Cloud, Hybrid Cloud, Private Cloud, and Edge (Beta). For pricing information, users are directed to explore the cloud offerings or talk to the team.