MemPalace

What is MemPalace?

MemPalace is an open-source memory system for AI assistants that stores conversation history word for word rather than summarizing or paraphrasing it. Content is organized in a structured index where people and projects become wings, topics become rooms, and the original text sits in drawers, so a search can be scoped to one area instead of scanning a flat pile of notes. Retrieval is semantic, uses ChromaDB by default, and the backend is pluggable so alternatives can be dropped in. Everything runs on your own machine and nothing leaves it unless you opt in, which makes it a fit for developers who use coding agents such as Claude Code and want recall without sending history to a hosted service. You can install it with uv, pipx, pip or a Docker image, wire it to an MCP client, or let an agent run the guided setup skills. The project is free, and the maintainers warn that only the GitHub repository, the PyPI package and mempalaceofficial.com are official.

Use cases

• Recall past decisions from coding agent sessions
• Search old conversations by meaning
• Keep AI memory on your own machine
• Mine project folders and transcripts into memory
• Give Claude Code a searchable history over MCP
• Scope searches to one project or topic

Who it's for

• Developers
• Claude Code users
• Privacy-minded AI users

Why use MemPalace

• Because text is stored verbatim, search returns what was actually said rather than a lossy summary.
• Wings and rooms let you limit a query to one person, project or topic.
• Data stays local by default, which suits private code and client conversations.
• A pluggable backend means you are not locked to one vector store.
• Agent-guided setup can detect your system, install the package and configure MCP for you.

MemPalace features

• Verbatim storage with no summarizing or paraphrasing
• Semantic search over stored conversations
• Wings, rooms and drawers structure for scoped searches
• Pluggable retrieval backend with ChromaDB as the default
• MCP server for connecting to coding agents
• Three agent skills for setup, recall and task delegation
• Multi-arch Docker image for the MCP server and CLI
• Mining of project folders and transcripts
• Private local palace, shared hub or client modes
• Optional weekly stable-release checks that never auto-install

MemPalace integrations

• Claude Code
• ChromaDB
• PyPI
• Docker
• uv
• pipx

How to get started with MemPalace

1. Install the CLI with uv tool install mempalace, pipx or pip, or pull the Docker image.
2. Run mempalace init on a project folder to create your palace.
3. Mine conversations, transcripts or project files so they are indexed as wings, rooms and drawers.
4. Connect MemPalace to your agent, for example Claude Code, as an MCP stdio server.
5. Ask the agent to search memory, or run mempalace search from the command line.

MemPalace limitations

• Requires a Python install or Docker image and some command-line setup
• Claude Code sessions expire in 30 days unless auto-save hooks are wired
• Native Termux installation on Android is not supported
• The first embedding run downloads a model of about 80 MB to 300 MB

MemPalace pricing

MemPalace is open source, so you can use and self-host it for free.

Prices can change and may vary by country or exclude tax. Check the official pricing page.

MemPalace alternatives

The main MemPalace alternatives listed on Debuly are Mem0, Zep, Letta, Supermemory, Memori and Cognee. The table compares pricing and platforms; order is A–Z.

1. 

   Cognee

   Open-source memory engine for AI agents that combines vector search, knowledge graphs and relational storage, with a hosted cloud.

   Developer platformsAutomation and agentsResearch and search
   FreemiumVisit Cognee (official site)
2. 

   Letta

   Platform, formerly MemGPT, for building stateful AI agents with persistent memory, from a terminal app, SDK or API.

   Developer platformsAutomation and agentsCoding
   FreemiumVisit Letta (official site)
3. 

   Mem0

   Memory layer for AI agents and apps, available as an open-source package or a managed platform that keeps user context across sessions.

   Developer platformsAutomation and agentsCoding
   FreemiumVisit Mem0 (official site)
4. 

   Memori

   Memory infrastructure for LLM apps and agents that captures conversations and tool traces as structured memories, in the cloud or on your own database.

   Developer platformsAutomation and agentsCoding
   FreemiumVisit Memori (official site)
5. 

   Supermemory

   Memory and context API for AI agents with user profiles, a memory graph, retrieval and connectors, billed by usage.

   Developer platformsAutomation and agentsCoding
   FreemiumVisit Supermemory (official site)
6. 

   Zep

   Managed memory and context platform for AI agents, built on the Graphiti knowledge graph, with Python, TypeScript and Go SDKs.

   Developer platformsAutomation and agents
   FreemiumVisit Zep (official site)

MemPalace FAQ

Sources & verification

Facts on this page come from the official sources below, each with the date we read it. Written by Debuly; the vendor did not review this page.

Image credits

• Logo: from avatars.githubusercontent.com · fetched 24 Oct 2026 · Vendor trademark, used only to identify the product; removed on request
• Cover: none collected yet, so a generated tile is shown.

Spot an error? Report an issue with this listing. We also remove a logo or image on the owner's request.