Qdrant Vector Search Engine
Looking for a Rust Database? Qdrant Is Written in Rust
Qdrant is a vector search engine written in Rust from storage to API, with an official Rust client. Start a free cluster on Qdrant Cloud with no credit card.
TRUSTED BY ENTERPRISE TEAMS
Written in Rust
Need a Database Written in Rust? Qdrant Is Rust From Storage to API
Qdrant is a vector search engine. Rust gives it memory safety and precise control of memory, so it stays fast and reliable under high load.

Keep latency steady under load.
Rust frees memory as soon as a value goes out of scope, so the engine keeps a low memory profile and steady response times as traffic grows.
Run parallel work safely.
Rust’s compiler rules out data races, which lets Qdrant parallelize indexing and search across cores.
Ship a small image.
The Qdrant Docker image is about 74 MB compressed on Docker Hub.
Rust client
Call It From Rust With the Official Client
Qdrant ships an official Rust client crate alongside Python, Go, Java, .NET, and JavaScript clients, over gRPC or REST.
Add one crate.
Run cargo add qdrant-client and connect over gRPC on port 6334 or REST on port 6333.
Load data at full speed.
The docs recommend the Rust client for bulk uploads as the fastest client library available for Qdrant.
Keep the payload with the vector.
Store a JSON payload next to each vector, filter on it during the search, and get it back with every result.
Deployment
Deploy the Rust Engine Yourself or Let Qdrant Cloud Run It
Start from a binary or a container.
Qdrant compiles to a single binary executable and ships as a Docker image.
Hand the cluster to Qdrant Cloud.
Qdrant Cloud runs the same engine as a managed service on AWS, GCP, or Azure, with upgrades, backups, and scaling handled for you.
Embed it with Qdrant Edge.
The qdrant-edge crate, in beta, runs the engine inside your application process, with a Qdrant server optional for backups and heavier indexing.
Results
Why Enterprise Teams Picked the Rust Engine
Bayer
chose Qdrant because it was open source and quick to test, latency was strong, and it was written in Rust. Today a four-node Hybrid Cloud cluster holds about 135 million points for 116,000 employees.
Read the Bayer case studyDeutsche Telekom
picked Qdrant for its simplicity, Rust-based efficiency, and memory management. Its agent platform now processes over 2 million conversations across three countries, and a new agent now takes 2 days to build, down from 15.
Read the Deutsche Telekom case studyGetting started
From Sign-Up to Your First Query in Rust
Create a free cluster.
Sign up for Qdrant Cloud with no credit card.
Add the client.
Run cargo add qdrant-client and connect with your cluster URL and API key.
Upsert and search.
Create a collection, upsert your vectors with their payloads, and run your first query from Rust.
