HarperDB
One Runtime for Apps, Agents, and Data.
Data Retrieval Speed
~1ms
LLM Call Reduction
Up to 85% fewer at scale
Award
2026 Data Breakthrough - Open Source Data Platform of the Year
About HarperDB
HarperDB provides a unified runtime platform designed to simplify modern application development by reducing architectural complexity. It integrates a database, cache, application logic, real-time pub/sub messaging, vector search, and blob storage into one high-performance process. This approach eliminates the need for separate services and integration overhead. HarperDB is built for developers and AI systems, offering a declarative surface that AI coding tools can reason with directly. It is suitable for a wide range of use cases, including real-time backends, edge and IoT systems, event-driven architectures, and distributed systems requiring low latency and high performance.
Unified Runtime Features
Database
High-performance database integrated directly into the runtime.
Caching
Built-in caching with sub-millisecond response times, no external cache needed.
Real-Time Messaging
Native real-time pub/sub over SSE, WebSocket, and MQTT without a separate broker.
Vector Search
Supports vector search alongside structured data in the same runtime.
Blob Storage
Native blob storage served from the same API endpoint as structured data.
Developer & AI Experience
Api Generation
Schema-driven REST API generation from GraphQL definitions, no controllers or ORM required.
Ai-Ready
Provides a declarative full-stack surface that AI systems (like Claude Code, Cursor, Codex) can reason with directly.
Deployment
Open source for local/self-hosted deployment, with global deployment available via Harper Fabric.
Operating System
Supports Linux, macOS, and Windows with a Node.js runtime.