memgraph
High-performance open-source in-memory graph database for GraphRAG, AI memory, agentic AI, and real-time graph analytics. Cypher-compatible, built in C++.
What is memgraph?
High-performance open-source in-memory graph database for GraphRAG, AI memory, agentic AI, and real-time graph analytics. Cypher-compatible, built in C++.
Overview of memgraph
What it is
Memgraph is a high-performance, in-memory graph database built in C/C++. It is designed for real-time AI context and graph analytics, serving as the graph engine for GraphRAG pipelines, AI memory systems, and agentic workflows. Memgraph provides sub-millisecond multi-hop traversals and combines built-in text and vector indexes with full graph traversal in a single query layer. It is fully compatible with Neo4j’s Cypher query language, ACID-compliant, and supports high availability.
Key features
- Built-in vector indexes – Enable hybrid graph retrieval with similarity search in a single query, alongside text and geospatial indexes.
- MAGE algorithm library – 40+ graph algorithms in C++, Python, and CUDA, including PageRank, community detection, GNN-based link prediction, and native ML.
- Atomic GraphRAP – Pivot search, graph expansion, ranking, and prompt assembly expressed as a single Cypher query.
- In-memory C/C++ engine – Sub-millisecond traversals with benchmarked performance.
- Streaming support – Ingest from Kafka, Pulsar, and RedPanda with dynamic graph algorithms that react to changes in real time.
- High availability – Raft-based coordination with automatic failover.
- Multi-tenancy – Isolated databases with per-tenant role assignments and fine-grained access control.
Use cases
Memgraph is used for GraphRAG pipelines, AI memory systems, agentic AI workflows, and real-time graph analytics. Specific applications include fraud detection, network analysis, infrastructure monitoring, and any operational workload requiring structured, connected context alongside semantic vector or text search.
Getting started
To deploy Memgraph on Kubernetes using Helm:helm repo add memgraph https://memgraph.github.io/helm-chartshelm install my-memgraph memgraph/memgraphBinaries and Docker images are available on the Memgraph Download Hub. For full installation instructions, see the repository README.
License and pricing
Memgraph Community is available under the BSL license. Memgraph Enterprise is available under the MEL license. The project is listed as open source in its metadata.
Based on the project's public README and repository metadata.
