Qdrant
A vector similarity search engine and vector database, written in Rust, with a production API for storing, searching and filtering vectors alongside their payload data.
Qdrant stores points, which pair a vector with an arbitrary payload, and serves similarity search over them with extended filtering support, making it suited to neural-network and semantic matching, faceted search, recommendation and related applications. It runs as a client-server deployment, most simply through a single Docker container, with official client libraries for Go, Rust, JavaScript/TypeScript, Python, .NET/C# and Java, plus community clients for Kotlin and PHP.
Qdrant Edge is a lightweight variant meant for edge devices and resource-constrained environments: rather than a separate server, it runs inside the application process, storing and querying data locally and optionally synchronizing with a Qdrant server, trading some server-side features for lower latency and offline operation. The project also publishes a set of agent skills that bring Qdrant-specific engineering guidance, such as choosing quantization, sharding, tenant isolation or hybrid search settings, directly into an AI coding assistant.
It is written in Rust, licensed under Apache-2.0, and also offered as a managed Qdrant Cloud service with a free tier.