FalkorDB
A self-hosted graph database for GraphRAG and AI agent memory, with Docker deployment, Cypher queries, vector search and a hosted cloud offering.
FalkorDB is a self-hosted graph database for developers building GraphRAG applications and AI agents that retrieve connected facts. It stores entities, relationships and their attributes, with Cypher queries and vector search for finding relevant information. Docker runs the database on your own hardware; a separate hosted cloud service is also available.
Graph traversal and similarity search can work together to retrieve context from knowledge graphs. Applications can use the database to store user preferences, conversation history or domain knowledge, then retrieve those connections for a model. The database supplies storage and queries; your application supplies the model and agent behavior.
FalkorDB's separate GraphRAG SDK adds workflows for structured and unstructured data, including automatic ontology generation. Those SDK capabilities are separate from the database engine.
The engine supports OpenCypher with query extensions and works through Redis clients. Dedicated clients cover Python, Java, Node.js, Rust, Go and C#. Its Rust implementation uses GraphBLAS sparse matrices and linear algebra for graph storage and traversal. Multi-graph support lets one instance hold separate graphs for different applications or tenants.
The source is available under SSPLv1. Providing its functionality as a service to third parties triggers the license's service-source requirements; it is not a permissive open-source license.
