Qdrant pockets $50M to push composable vector search

Open source vector database supplier Qdrant has raised $50 million in B-round funding, taking total funding to $87.5 million since its 2021 founding.

Its pitch is that vector search, detecting similarities between AI token vectors in a multidimensional search space, has changed. Originally it involved retrieving nearest neighbors from dense vector embeddings over relatively static datasets. The static aspect is disappearing as retrieval now runs within agent loops, executing thousands of queries per workflow across hybrid modalities, against data that changes continuously. The vector database has to scale its size and performance to cope.

André Zayarni, Qdrant CEO and co-founder, said: "Many vector databases were built to only store dense embeddings and return nearest neighbors. That's table stakes. Production AI systems need a search engine where every aspect of retrieval – how you index, how you score, how you filter, how you balance latency against precision – is a composable decision." 

Composable vector search means teams choose and combine retrieval capabilities at query time: dense vectors, sparse vectors, metadata filters, multi-vector representations, and custom scoring functions, with explicit control over how each affects relevance, latency, and cost. Qdrant says it rethinks every layer of retrieval – indexing, scoring, filtering, ranking – as composable primitives that engineers control directly. 

Whether a team optimizes for maximum accuracy, lowest latency, or cost efficiency at scale, Qdrant says the result is a search engine that adapts to the problem rather than forcing the problem to fit the tool. It argues that RAG pipelines, semantic search, and agentic reasoning depend on retrieval that holds up under sustained, production-scale pressure. Tools limited to single-vector dense similarity or architectures that layer vector search onto legacy indexing models are breaking under these conditions.

Qdrant's composable vector search is designed to operate wherever decisions are made; in the cloud, in hybrid and private (on-prem) environments, or at the edge.

The Series B funding was led by AVP, with participation from Bosch Ventures, Unusual Ventures, Spark Capital, and 42CAP. Warda Shaheen, General Partner of AVP, said: "With every infrastructure shift, we've seen purpose-built systems emerge and rapidly scale in fast-growing new markets, and we're seeing this pattern again with Qdrant."

Qdrant's open source project has passed 250 million downloads and 29,000 GitHub stars. It says its software delivers predictable, low-tail latency across cloud, hybrid, on-premises, and edge deployments. Learn more here.