Qdrant
Qdrant is a vector similarity search engine and vector database written in Rust. It provides a production-ready service with a convenient API to store, search, and manage points (vectors with payload). It is designed for developers building AI applications that require semantic search, recommendations, and other vector-based matching with extended filtering support.
✨ Key features
- Dense, sparse, and multi-vector search support
- Rich payload filtering with JSON conditions
- Hybrid search with fusion strategies (RRF, DBSF)
- Vector quantization reduces RAM usage by up to 97%
- Distributed deployment with sharding and replication
- Web UI for visual data exploration and management
🎯 Use cases
- Semantic text search over documents or product descriptions
- Similar image search for visual discovery applications
- Recommendation systems using positive and negative examples
- E-commerce product categorization with extreme classification
- Multitenant applications requiring scalable data partitioning
📦 Installation
🧰 Requirements: Docker is required for the quick start container deployment. No API keys needed for local setup.
To run Qdrant locally, use Docker:
docker run -p 6333:6333 qdrant/qdrant
This starts an insecure deployment without authentication. For production, refer to the security guide.
🚀 Usage
Connect to the server using a client library. Example with Python:
from qdrant_client import QdrantClient
client = QdrantClient(url="http://localhost:6333")
Then you can create collections, upsert points, and search.
⚠️ Good to know
The quick start command runs an insecure deployment without authentication, open to all network interfaces; production deployments should follow security guidelines.
❓ FAQ
What is Qdrant?
Qdrant is a vector similarity search engine and vector database that provides a production-ready service with an API to store, search, and manage points (vectors with payload).
How do I run Qdrant locally?
You can run Qdrant locally using Docker with the command: docker run -p 6333:6333 qdrant/qdrant.
What client libraries are available?
Official clients are available for Go, Rust, JavaScript/TypeScript, Python, .NET/C#, and Java. Community clients exist for Kotlin and PHP.
Does Qdrant support filtering on payload?
Yes, Qdrant allows attaching JSON payload to vectors and filtering on it using conditions like keyword matching, numeric ranges, geo-locations, and more.
📊 Repository
🤖 Overview, features, install steps and FAQ were generated from the project's README on Sep 4, 2026. Always check the original source before running commands.