Qdrant, the open-source vector search engine built in Rust, has closed a $50 million Series B round. AVP led the investment, with Bosch Ventures, Unusual Ventures, Spark Capital, and 42CAP joining as participants.
What it did make clear is that the capital will go toward accelerating what Qdrant calls composable vector search — a technical architecture that gives engineers direct control over how their AI systems retrieve, rank, and filter information at scale.
Vector search started as a fairly niche concern. The original use case was simple: given a dataset of dense embeddings, find the nearest neighbors. That worked fine for research environments and early demos. Production AI systems are a different matter entirely.
Modern AI applications run retrieval inside agent loops, firing off thousands of queries per workflow. They work across hybrid data types, handle continuously changing data, and feed results into RAG pipelines, semantic search engines, and agentic reasoning systems. The infrastructure underneath all of that needs to hold up under sustained, real-world load — and older approaches, whether single-vector dense similarity or vector search bolted onto legacy indexing systems, are struggling to keep pace.
"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. As an AI-native vector search engine designed for the latency, throughput, and reliability demands of production AI workloads, they're at the forefront of building the retrieval layer of the future that all advanced AI applications will depend on," said Warda Shaheen of AVP.
"In production AI applications, retrieving context-relevant information in real-time has become business-critical infrastructure," said Ingo Ramesohl, Managing Director of Bosch Ventures. "Qdrant's Rust-based architecture is exemplary of the deep tech innovations that will shape the next generation of powerful and trustworthy AI systems."
Qdrant plans to use the funding to establish composable vector search as the default approach for production AI retrieval. That means deepening the engine's capabilities across cloud, hybrid, on-premises, and edge deployments — wherever enterprises need to run AI at the infrastructure level.
The company is also investing in expanding its reach among both developers and enterprise teams. The goal is to make Qdrant the retrieval layer that serious AI applications are built on, not just a useful tool in an experimental stack. That involves continued development of the open-source project, growth across global markets, and building out the team to support it.
Qdrant was built from the ground up in Rust, chosen for its performance and memory safety. Every layer of the retrieval system — indexing, scoring, filtering, ranking — is designed as a composable primitive. Engineers don't have to accept opaque defaults. They choose and combine capabilities at query time: dense vectors, sparse vectors, metadata filters, multi-vector representations, and custom scoring functions, with explicit control over how each one affects relevance, latency, and cost.
"Many vector databases were built to only store dense embeddings and return nearest neighbors. That's table stakes," said André Zayarni, CEO and Co-Founder of Qdrant. "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. That's what we've built, that's what developers and the most sophisticated enterprises are looking for as they scale internal and external AI workloads, and this funding accelerates our ability to make it the standard."
The practical result is a search engine that adapts to the problem. Whether a team is optimizing for accuracy, low latency, or cost efficiency at scale, Qdrant exposes the controls to get there without requiring a full re-architecture as requirements change.
Qdrant's open-source project has surpassed 250 million downloads and 29,000 GitHub stars. The global developer community contributes improvements drawn from real production environments, which keeps the roadmap grounded in actual engineering challenges rather than theoretical ones.
Enterprise clients including Tripadvisor, HubSpot, OpenTable, Bazaarvoice, and Bosch run Qdrant in continuous production environments. The company has also received formal industry recognition: it was included in The Forrester Wave: Vector Databases, Q3 2024, GigaOm's Radar for Vector Databases v3 in 2025, and Sifted's 2025 B2B SaaS Rising 100.
Prior investors Unusual Ventures, Spark Capital, and 42CAP all returned for the Series B alongside new participants AVP and Bosch Ventures.
