← Discover
Discover · GitHub · Mempalace↑ +227 today

Add memory to AI models

MemPalace offers an open source memory system for artificial intelligence models. The system includes optimized tools to measure and improve the contextual memory performance of large language models.

Updates

  • September 16, 2026: Stars 58,752 → 59,089, latest release v3.10.0 (September 16, 2026).
  • August 31, 2026: Stars 58,584 → 58,752, latest release v3.9.0 (August 31, 2026).
  • August 24, 2026: Stars 58,390 → 58,584, latest release v3.8.0 (August 23, 2026).
  • August 15, 2026: Stars 58,311 → 58,390, latest release v3.7.1 (August 14, 2026).

What you get

  • It offers semantic search by keeping the conversation history as it is.
  • It creates an organized memory structure by categorizing people, projects and topics.
  • With its local operating principle, it allows data to be processed without leaving your machine.

Installation

installation with uv
uv tool install mempalace
mempalace init ~/projects/myapp
Installation with virtual environment
python -m venv .venv && source .venv/bin/activate
pip install mempalace

If you don't write code

🤖 Paste this into your AI agent (Claude Code · Codex · Antigravity)

I want to set up my local AI memory system using the MemPalace tool. Can you explain to me the basic steps I need to follow to create a structure where I can scan my project files and conversation history and perform semantic searches, and the process from installation to the first search?

Related dictionary terms

Who it is forFor users looking for a local and privacy-focused memory system for their AI models.
LicenseMIT

Links

TreScout did not build this tool · we found it in GitHub trends and wrote it up. This page describes the repository as of 2026-06-06: The star count and our text belong to that day, the repository may have changed since. Check the repository link for the current state. This page was machine-translated from the Turkish original · the Turkish version prevails.