SuperLocalMemory
World's first local-only AI memory to break 74% retrieval and 60% zero-LLM on LoCoMo. No cloud, no APIs, no data leaves your machine. Additionally, mode C (LLM/Cloud) - 87.7% LoCoMo. Research-backed. arXiv: 2603.14588
What is SuperLocalMemory?
SuperLocalMemory is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to world's first local-only ai memory to break 74% retrieval and 60% zero-llm on locomo. no cloud, no apis, no data leaves your machine. additionally, mode c (llm/cloud) - 87.7% locomo. research-backed. ...
World's first local-only AI memory to break 74% retrieval and 60% zero-LLM on LoCoMo. No cloud, no APIs, no data leaves your machine. Additionally, mode C (LLM/Cloud) - 87.7% LoCoMo. Research-backed. arXiv: 2603.14588
This server falls under the Knowledge & Memory category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- World's first local-only AI memory to break 74% retrieval an
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx superlocalmemoryConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use SuperLocalMemory
SuperLocalMemory is a privacy-first persistent AI memory MCP server that stores and retrieves all conversation context entirely on your local machine — no cloud, no external APIs. It uses a 5-channel hybrid retrieval system (semantic, BM25, entity graph, temporal, and Hopfield networks) to achieve 74.8% accuracy on the LoCoMo benchmark, outperforming cloud-based memory solutions. Developers, researchers, and privacy-conscious users integrate it with Claude, Cursor, VS Code, and 17+ other AI tools to give their AI assistants continuous, accurate memory across sessions.
Prerequisites
- Node.js 14+ (for npm installation) or Python 3.11+ (for pip installation)
- No GPU required — the math engine runs CPU-only
- Approximately 700MB of disk space for the embedding model and dependencies
- Optional: Ollama installed locally for Mode B (enhanced answers with local LLM)
- An MCP-compatible client such as Claude Desktop, Cursor, or VS Code
Install SuperLocalMemory via npm
Install the package globally using npm. This is the recommended method as it provides the 'slm' CLI command.
npm install -g superlocalmemoryRun the setup wizard
Run the setup command to choose your operating mode. Mode A is fully local (zero cloud), Mode B uses a local Ollama LLM for better answers, Mode C allows optional cloud LLM for maximum accuracy.
slm setupVerify the installation
Run the doctor command to confirm all components are working correctly before connecting to your AI client.
slm doctorPre-download the embedding model (optional but recommended)
Download the embedding model (~500MB) in advance so memory operations don't pause on first use.
slm warmupConfigure your MCP client
Add SuperLocalMemory to your MCP client using either HTTP transport (recommended for v3.6.7+, one shared daemon) or stdio transport (universal, spawns a subprocess per connection).
Start the MCP server
Launch the MCP server manually if using stdio transport, or let 'slm mcp' run as a daemon for HTTP transport.
slm mcpSuperLocalMemory Examples
Client configuration
Add one of these to your claude_desktop_config.json. Use HTTP transport for the recommended shared daemon approach (v3.6.7+) or stdio for universal compatibility.
{
"mcpServers": {
"superlocalmemory": {
"type": "http",
"url": "http://127.0.0.1:8765/mcp/"
}
}
}
// OR for stdio transport:
{
"mcpServers": {
"superlocalmemory": {
"command": "slm",
"args": ["mcp"]
}
}
}Prompts to try
Once connected, your AI assistant can store and retrieve memory across sessions using these example prompts.
- "Remember that our database uses PostgreSQL 16 on port 5432 with the schema named 'production'"
- "What database are we using in this project?"
- "Remember that Alice is the lead engineer and prefers code reviews on Tuesdays"
- "What do you know about our team members?"
- "Forget the note about the old Redis cache configuration from last week"Troubleshooting SuperLocalMemory
'slm' command not found after npm install
The npm global bin directory may not be on your PATH. Run 'npm config get prefix' to find where npm installs global binaries, then add the 'bin' subdirectory to your PATH in ~/.bashrc or ~/.zshrc.
Memory recall returns empty or inaccurate results
Run 'slm doctor' to check that the embedding model is loaded correctly. If you skipped 'slm warmup', the first recall may be slow while the model downloads. Check 'slm health' for math layer diagnostics.
MCP client cannot connect to the HTTP endpoint
Ensure the daemon is running with 'slm mcp' before connecting. Verify the port with 'slm status'. If port 8765 is in use, check for conflicting processes and use 'slm reap' to clean up orphaned instances.
Frequently Asked Questions about SuperLocalMemory
What is SuperLocalMemory?
SuperLocalMemory is a Model Context Protocol (MCP) server that world's first local-only ai memory to break 74% retrieval and 60% zero-llm on locomo. no cloud, no apis, no data leaves your machine. additionally, mode c (llm/cloud) - 87.7% locomo. research-backed. arxiv: 2603.14588 It connects AI assistants to external tools and data sources through a standardized interface.
How do I install SuperLocalMemory?
Follow the installation instructions on the SuperLocalMemory GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with SuperLocalMemory?
SuperLocalMemory works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is SuperLocalMemory free to use?
Yes, SuperLocalMemory is open source and available under the AGPL-3.0 license. You can use it freely in both personal and commercial projects.
SuperLocalMemory Alternatives — Similar Knowledge & Memory Servers
Looking for alternatives to SuperLocalMemory? Here are other popular knowledge & memory servers you can use with Claude, Cursor, and VS Code.
MemPalace
★ 52.6kA local AI memory system that stores all conversations verbatim and organizes them into navigable structures. It provides 19 MCP tools for AI assistants to search and retrieve past decisions, debugging sessions, and architecture debates automatically
Kratos
★ 25.7k🏛️ Memory System for AI Coding Tools - Never explain your codebase again. MCP server with perfect project isolation, 95.8% context accuracy, and the Four Pillars Framework.
Context Mode
★ 15.4kAn MCP server that preserves LLM context by intercepting large data outputs and returning only concise summaries or relevant sections. It enables efficient sandboxed code execution, file processing, and documentation indexing across multiple programm
Memu
★ 13.7kMemory for 24/7 proactive agents like OpenClaw.
MemOS
★ 9.3kMemOS (Memory Operating System) is a memory management operating system designed for AI applications. Its goal is: to enable your AI system to have long-term memory like a human, not only remembering what users have said but also actively invoking, u
Everos
★ 5.4kBuild, evaluate, and integrate long-term memory for self-evolving agents.
Browse More Knowledge & Memory MCP Servers
Explore all knowledge & memory servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up SuperLocalMemory in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use SuperLocalMemory?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.