Qdrant is an open-source vector search engine built entirely in Rust, founded in 2021 by Andre Zayarni and Andrey Vasnetsov. Headquartered in Berlin, Germany, the company designed Qdrant from the ground up for production-grade AI retrieval at scale, prioritizing throughput, memory efficiency, and predictable latency under real-world load rather than benchmark numbers in isolation.
The engine supports dense, sparse, and multivector search, advanced metadata filtering, hybrid queries that combine vector similarity with payload predicates, and reranking. Recent releases added scalar and binary quantization for dramatic memory savings, on-disk vectors for cost-efficient large indexes, multitenancy primitives, and a distributed mode for horizontal scaling. The Apache 2.0 licensed core can be deployed on-prem, hybrid, edge, or as Qdrant Cloud across AWS, GCP, and Azure.
Qdrant raised a $50M Series B in March 2026 led by AVP (formerly AXA Venture Partners) with participation from Bosch Ventures, Unusual Ventures, Spark Capital, and 42CAP. The round nearly doubled the $28M Series A from 2024 and brought total funding to approximately $88M. Enterprises including Tripadvisor, HubSpot, OpenTable, Bazaarvoice, Bosch, and Cohere run Qdrant in production where continuous vector workloads are core to the product.
Pricing for Qdrant Cloud starts with a free 1 GB cluster, then scales through paid Cloud, Hybrid Cloud (control plane in Qdrant Cloud, data plane in the customer's own cloud account), and Private Cloud deployments billed on cluster size and storage. The open-source core is free to self-host with no feature gating.
Qdrant competes head-on with Pinecone, Weaviate, Milvus, and pgvector. Its differentiation is Rust-driven performance, a single-binary deployment story that ops teams appreciate, and a deliberately composable architecture that lets engineers mix dense, sparse, and reranking models without sacrificing speed.