​

HelixDB Overview

HelixDB is a graph-vector database designed for modern applications that require complex relationship modeling and traversal. It combines the power of graph data structures for explicit relationships with vector embeddings for semantic search.

​

Key Features

• High Performance: Optimized for fast traversals and insertions
• Native Graph Storage: Purpose-built storage engine optimized for graph operations
• Native Vector Storage: Purpose-built storage engine optimized for vector operations
• Type Safety: Explicit type definitions to ensure data integrity
• Developer-Friendly: Simple setup and intuitive APIs

​

Architecture

​

Data Model

HelixDB follows the property graph model, which consists of:

• Nodes: Represent entities with unique identifiers and properties
• Vectors: Represent entities with unique identifiers and vector embeddings
• Edges: Represent relationships between nodes and/or vectors, with directionality and properties
• Properties: Key-value pairs attached to both nodes, vectors and edges
• Labels: Categorize nodes and edges by type

This model allows for rich representation of complex domains while maintaining query performance.