# memgraph/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++.

Repository: https://github.com/memgraph/memgraph
Canonical: https://ross.abutalabs.com/products/memgraph
Homepage: https://memgraph.com
Language: C++
License: NOASSERTION
License Family: other
Topics: graph-algorithms, graph-database, graph-analysis, kafka, cypher, graph, opencypher, ai, graphrag, ai-agents, ai-memory, context-graph, enterprise-context, streaming, nosql
Last push: 2026-08-26T23:08:13+00:00

## Health v2 (maintenance only)
Score: 97/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 93, longevity 100
- inputs: {"age_days": 2172, "days_push": 7, "days_rel": 49, "gap_med": 21, "n_releases_24m": 28}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 4364, forks 259 (observed 2026-08-28T04:08:46.592735+00:00)

## What it is
Memgraph is a high-performance, in-memory graph database written in C++ that is Cypher-compatible and ACID-compliant, with built-in vector, text, and geospatial indexes. It targets GraphRAG pipelines, AI memory systems, agentic AI reasoning, and real-time graph analytics such as fraud detection and network analysis.

## Use cases
- run graphrag pipelines with multi-hop traversal and vector search in one query
- store long-term memory for llm agents as a connected graph
- detect fraud in real time by analyzing entity relationships
- migrate from neo4j to a faster in-memory graph database
- run pagerank and community detection on streaming data from kafka
- build a knowledge graph from csv or sql data for ai context

## When to choose
- you need sub-millisecond multi-hop graph traversals for AI or analytics workloads
- you want graph traversal and vector similarity search in a single atomic query
- you already use Cypher and want Neo4j compatibility with better performance
- you process streaming data from Kafka, Pulsar, or Redpanda into a live graph

## When to avoid
- you need a fully open-source license for commercial use - Memgraph uses BSL/MEL source-available terms
- your dataset exceeds available RAM and you need disk-first storage
- you need a lightweight embedded graph store rather than a server deployment

## Facets
- artifact type: service
- maturity: stable
- function: database, vector-database, search-engine, streaming, rag, machine-learning, analytics
- domain: databases, large-language-models, analytics, big-data
- platform: self-hosted, cloud, cpp, python
- tags: graph-database, cypher, opencypher, in-memory, graphrag, ai-memory, agentic-ai, mage, neo4j-compatible, kafka, fraud-detection, graph-algorithms, bsl-license, ai-agents, retrieval-augmented-generation, linux, docker, kubernetes

## Member repositories
- memgraph/memgraph (main) score 97

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:08:46.592735+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:21:25.751713+00:00, confidence not recorded.
  - readme: https://github.com/memgraph/memgraph (fetched 2026-08-28T04:08:46.592735+00:00, sha dc25a51a1a1f)
  - homepage: https://memgraph.com (fetched 2026-08-29T09:09:56.405976+00:00, sha 76bfae61a0e3)
  - site_page: https://memgraph.com/docs (fetched 2026-08-29T09:09:56.419681+00:00, sha 585aadc04d2b)
  - site_page: https://memgraph.com/docs/getting-started (fetched 2026-08-29T09:09:56.421992+00:00, sha 7436b6d9d366)
  - site_page: https://memgraph.com/about-us (fetched 2026-08-29T09:09:56.424116+00:00, sha d7d982c4f999)
  - site_page: https://memgraph.com/tools (fetched 2026-08-29T09:09:56.415308+00:00, sha 38b45be95ca0)
  - site_page: https://memgraph.com/pricing (fetched 2026-08-29T09:09:56.417619+00:00, sha 91f97b80bd46)
- Data as of 2026-08-30T08:39:29.467469+00:00.