Multiple ways to install MemoryGraph depending on your needs.
Prerequisites
- Python 3.10+ - Required for the MCP server
- pip - Python package installer
- pipx (recommended) - Isolated Python app installer
# Check Python version
python3 --version # Should be 3.10+
# Install pipx if you don't have it
pip install --user pipx
pipx ensurepath Installation Methods
Method 1: pipx (Recommended)
pipx installs Python applications in isolated environments and handles PATH automatically.
# Basic installation (SQLite default)
pipx install memorygraphMCP
# With FalkorDBLite (embedded graph database)
pipx install "memorygraphMCP[falkordblite]"
# With FalkorDB client (requires FalkorDB server)
pipx install "memorygraphMCP[falkordb]"
# With Neo4j support (enterprise)
pipx install "memorygraphMCP[neo4j]" Method 2: pip
Standard pip install (ensure Python bin directory is in PATH).
# Basic installation
pip install memorygraphMCP
# With extras
pip install "memorygraphMCP[intelligence]" Method 3: From Source
For development or latest unreleased features.
# Clone the repository
git clone https://github.com/gregorydickson/memory-graph.git
cd memory-graph
# Install in development mode
pip install -e ".[dev,intelligence]" Verify Installation
# Check if memorygraph is available
which memorygraph
memorygraph --version
# Run health check
memorygraph --health Backend Options
| Extra | Backend | Use Case |
|---|---|---|
| (none) | SQLite (default) | Simple local storage, zero config |
falkordblite | FalkorDBLite (embedded) | Native graph queries without a server |
falkordb | FalkorDB (client-server) | High-performance production deployments |
neo4j | Neo4j | Enterprise features, large-scale |
dev | Development deps | Contributing to MemoryGraph |
Next Steps
After installation, configure MemoryGraph with your AI assistant: