mempalace
The best-benchmarked open-source AI memory system. And it's free.
What is mempalace?
The best-benchmarked open-source AI memory system. And it's free.
Overview of mempalace
What it is
MemPalace is a local-first AI memory system that stores conversation history as verbatim text and retrieves it with semantic search. It does not summarize, extract, or paraphrase. The index is structured: people and projects become wings, topics become rooms, and original content lives in drawers, so searches can be scoped rather than run against a flat corpus. The retrieval layer is pluggable; the current default is ChromaDB. Nothing leaves your machine unless you opt in.
Key features
- Verbatim storage: stores original text without summarization or paraphrasing.
- Structured index: uses wings, rooms, and drawers to scope searches.
- Pluggable backends: supports ChromaDB, SQLite exact, Rust exact, Milvus, Qdrant, and pgvector.
- Local-first: all data stays on your machine unless you opt in.
- CLI and MCP server: ships a command-line interface and an MCP server for integration with AI agents.
- Docker support: multi-arch container image (amd64 + arm64) for running without a local Python toolchain.
Use cases
- Storing and retrieving conversation history for AI agents.
- Providing semantic search over verbatim text without external API calls.
- Scoped memory retrieval for specific projects, people, or topics.
- Running a shared memory hub for team collaboration.
Getting started
Install the MemPalace skills for agent-guided setup: bash npx skills add MemPalace/mempalace For direct CLI setup, use uv or pipx: bash uv tool install mempalace mempalace init ~/projects/myapp Or install with pip inside a virtual environment: bash python -m venv .venv && source .venv/bin/activate pip install mempalace Docker image: bash docker pull ghcr.io/mempalace/mempalace:latest
License and pricing
The license is MIT. The software is free and open source.
Based on the project's public README and repository metadata.