qdrant/qdrant

Project overview

A vector similarity search engine that stores, searches, and manages dense, sparse, and multi vectors with attached JSON payloads for AI and semantic matching applications. It is written in Rust and provides extended filtering, distributed deployment, and built-in quantization to manage RAM usage.

Project type

AI Search · Vector Database · Infrastructure

Use cases

Knowledge Q&A · Search & Research · Data Analysis

Deployment

Docker / Docker Compose

License

Apache-2.0

Best for

• Developers, data teams, and AI engineers who need to store and search vectors with JSON payloads for AI search, enterprise RAG, or AI data analysis.
• Teams that require horizontal scaling, sharding, and replication for large vector collections.

Key capabilities

• Supports dense vectors for semantic similarity, sparse vectors for full-text search, and multivector search for objects with multiple embeddings or late interaction models like ColBERT.
• Allows attaching any JSON payload to vectors and filtering on it using keyword matching, full-text, numeric ranges, and geo-locations.
• Combines multiple vectors in a single query to merge semantic understanding and keyword precision via configurable fusion strategies such as RRF and DBSF.
• Built-in quantization reduces RAM usage by up to 97% and allows tuning the trade-off between search speed and precision.
• Enables horizontal scaling with sharding and replication, and allows updating or resizing collections with zero downtime.
• Provides a visual way to interact with data, monitor deployment health, explore collections, manage data, and interact with the REST API.
• Provides a REST API with an OpenAPI 3.0 specification and a gRPC interface for production-tier searches.
• A lightweight version for edge devices and resource-constrained environments that runs inside the application process with local data and synchronization.

Limitations and risks

• Starts an insecure deployment without authentication by default, exposing the system to all network interfaces.
• Running the default Docker container exposes an insecure deployment open to all network interfaces without authentication.

Getting started

• Run the Docker command `docker run -p 6333:6333 qdrant/qdrant`. The setup is rated as easy due to the available Docker run command and clear quickstart documentation.

Evidence and sources

• GitHub project description: Qdrant - High-performance, massive-scale Vector Database and Vector Search Engine for the next generation of AI. Also available in the cloud https://cloud.qdrant.io/
• README: Qdrant is written in Rust 🦀, which makes it fast and reliable even under high load.
• README: To experience the full power of Qdrant locally, run the container with this command: ```bash docker run -p 6333:6333 qdrant/qdrant ``` Note that this starts an insecure deployment…
• README: It provides a production-ready service with a convenient API to store, search, and manage points—vectors with an additional payload.
• README: Qdrant supports dense vectors for semantic similarity, sparse vectors for full-text search, and multivector search for objects with multiple embeddings or late interaction models…

Reviewed at: 2026-07-17T09:22:35Z