Hanzo Vector

High-performance vector database for AI embeddings and similarity search.

Overview

Hanzo Vector is a production-ready vector database optimized for AI workloads. It provides efficient storage and retrieval of high-dimensional vectors, enabling lightning-fast similarity search for embeddings generated by large language models, image encoders, and other AI systems.

Features

• High Performance: Optimized HNSW indexing for fast approximate nearest neighbor search
• Scalability: Horizontal scaling with distributed architecture
• Flexible Filtering: Combine vector search with payload filtering
• Multiple Distance Metrics: Cosine, Euclidean, Dot Product
• REST & gRPC APIs: Easy integration with any language
• Cloud Native: Kubernetes-ready with Helm charts

Quick Start

docker run -p 6333:6333 -p 6334:6334 hanzo/vector

Documentation

See the documentation for detailed guides and API reference.

Source

The engine lives on the master branch. This branch carries the project description only.

License

Apache-2.0 - see LICENSE for details.

Hanzo Vector is derived from Qdrant, which is Apache-2.0. See NOTICE for attribution.