FalkorDB
- About
- Low-latency graph database for GraphRAG and agent memory. Open-source alternative to Neo4j, Amazon Neptune and TigerGraph.
- Catalogue
- No. 398
- Category
- Development
- GitHub stars
- ★ 0
Overview
FalkorDB is a multi-tenant property graph database built for low latency, aimed at knowledge graphs for LLMs, agent memory, cloud security and fraud detection. It represents graphs as sparse matrices and runs queries with linear algebra.
What it does well
FalkorDB supports the property graph model and the OpenCypher query language with its own extensions. You can start it with Docker, explore it in a browser UI and use client libraries in several languages, including Python.
Why it’s in the collection
FalkorDB is a fast way to give AI apps a knowledge graph, with an approach to graph queries unlike most databases. It's under the Server Side Public License.
License
FalkorDB uses the Server Side Public License (SSPL) v1, a source-available license. You can use, modify and self-host it, but if you offer FalkorDB to others as a service you must release the source of your whole service stack.
On GitHub
- Stars
- 0
- Forks
- 0
- Commits
- 0
- Repository
- FalkorDB/FalkorDB
Languages
- Rust 52.3%
- Python 26.5%
- Gherkin 12.5%
- Other 8.7%
Neo4j
Amazon Neptune
TigerGraph
