Qdrant
Apache-2.0 vector database and similarity-search engine written in Rust, run on the user's own infrastructure.
What it does
- Vector search
- Knowledge retrieval
Features 9
- Advanced compressionScalar, Product and Binary Quantization reduce memory usage and improve search performance for high-dimensional vectors.
- MultitenancySegments a single collection for efficient retrieval, data isolation and privacy.
- Payloads and advanced filteringVector payloads support many data types and query conditions such as string matching, numerical ranges and geo-locations.
- Sparse vector supportHandles sparse vectors efficiently for text retrieval and memory-efficient high-dimensional data.
- OpenAPI v3 specificationOffers an OpenAPI v3 specification for generating client libraries in almost any programming language.
- Memory maps and IO UringProvides on-disk storage options and low-level hardware optimization.
- Hybrid searchCombines semantic and lexical search in one retrieval.
- FastEmbedA library for generating embeddings on the client side before sending vectors to Qdrant.
- gRPC interfaceExposes a gRPC method for every REST endpoint.
How it compares
Sources
- Pricing
-
Official documentation · 18 Sep 2026
Licence read from the GitHub API's own license field: Apache-2.0.
- Description
-
Official documentation · 18 Sep 2026
Product page, fetched 2026-09-18: 'Qdrant is the most advanced vector search with highest RPS, minimal latency, fast indexing, high control with accuracy.' The repository describes it as a 'High-performance, massive-scale Vector Database and Vector Search Engine'.
- Sold within
-
Official documentation · 18 Sep 2026
Downloaded and run on its own under the open licence.
- URL
-
Official documentation · 18 Sep 2026
Fetched 2026-09-18, 200; first item in Qdrant's own Products nav.
- Interfaces
-
Official documentation · 22 Oct 2026
Qdrant docs Interfaces page, read 2026-10-22: "Qdrant supports these “official” clients." (Python, JavaScript / Typescript, Rust, Go, .NET, Java) and "All interaction with Qdrant takes place via the REST API." The Qdrant MCP Server appears only as a docs tutorial entry, not tied to this product, so mcp-server is not added.
Also from Qdrant
-
Qdrant Cloud Data service
Managed Qdrant clusters on AWS, Google Cloud and Azure, with a free single-node tier and consumption billing above it.
-
Qdrant Hybrid Cloud Data service
Qdrant's managed control plane over clusters a customer runs in their own Kubernetes, in any cloud, on-premises or at the edge.
-
Qdrant Private Cloud Data service
Fully on-premise Qdrant deployment for buyers who cannot let a vendor-run control plane touch their environment.
-
Qdrant Cloud Inference API service
Embedding generation run inside Qdrant Cloud so text or images are vectorised and searched in a single API call.
-
Qdrant Edge Data service
In-process vector search engine for embedded devices, autonomous systems and mobile agents, in beta.
-
Qdrant Serverless Data service
Announced serverless edition of Qdrant Cloud in which a collection is created and queried with no cluster to size or scale.
Something wrong here? Send a correction — quote this product id: qdrant.