Graph-MIND MCP server

<!-- mcp-name: io.github.goyohan0611-png/graph-mind -->

Graph-MIND

Your AI forgets everything when the session ends. This one remembers, and it never calls an LLM to do it.

88.8% on LongMemEval. Zero model tokens spent saving memory. Runs on your own PC.

[![][license-shield]][license-link] [![][python-shield]][python-link] [![][platform-shield]][platform-link] [![][longmemeval-shield]][benchmarks-link] [![][ci-shield]][ci-link] [![][pypi-shield]][pypi-link]

Decide something in Claude Code on Monday. Ask Codex on Wednesday. It knows.

A real run, not a mock-up: python demo/record_demo.py records it from a fresh store.

The part that sounds wrong

Same 40 questions, same answer model, same prompt, same official judge, pre-registered before anyone looked:

Mem0 runs an LLM over your conversations to decide what to keep. Graph-MIND keeps them word for word and finds the right part when you ask. It is cheaper, and in this test it was also more accurate.

Don't believe it? Good. Every question, answer and judge verdict is in research/runs/, and REPORT.md shows how to rerun it, including the runs that went wrong. Or just try it on your own conversations:

pip install graph-mind-memory
graph-mind-install          # connects Claude Code, Codex, Claude Desktop, Cursor, VS Code, ...

[!NOTE]

Alpha. Windows: used daily, with full end-to-end checks (fresh install, two PCs, real

Claude and Codex). macOS and Linux: tests and a clean install from PyPI pass in CI on every

release and every day, but nobody has used them day to day yet; consider them experimental.

What it is

Graph-MIND records every conversation you have with Claude Code, Codex and Claude Desktop, word for word, in a store on your own PC. When a question depends on the past, the model you are using calls Graph-MIND over MCP and gets back a few thousand tokens of the conversations that answer it.

with keyword search. Korean and English both work.

• Nothing is summarised or rewritten. Saving is a database write. No model call, no tokens.
• Search is hybrid. Your words are embedded on your PC (multilingual MiniLM, no API) and fused

stores the turns and embeds them as they arrive.

• Capture is automatic. A background service reads each app's local transcripts, masks secrets,

others (see below).

• One memory across your PCs. One sentence to the AI on the first PC, one command on the

leaves them is what your AI app already sends: the recalled turns go to your model's provider with your question.

• No cloud of its own. No account, no telemetry; the memory stays on your machines. What

Benchmarks

All numbers below come from files in this repository: the pre-registrations, each question's answer, and the judge's verdict on it, under research/runs/. The method is in REPORT.md, including the failures and the corrections.

LongMemEvalS: answer accuracy, 500 questions.** The answer model is gpt-5-mini; the judge is the official gpt-4o-2024-08-06.

Shortened. The full README is on GitHub.