LibreAI / Agents / Memoh
// agents Β· felinics/Memoh

Memoh

Multi-agent platform where every agent gets its own computer and long-term memory.

πŸ’Ž Hidden gemβ˜… 2.6k starsβ‘‚ 253 forksAGPL-3.0updated 2026-10
β–¦ visual β€” this project's README ships no embeddable screenshot,
so here's a placeholder. The guide below is complete regardless.
// project facts

Project facts

Memoh is an open-source agents project by felinics (felinics/Memoh on GitHub). It has 2,639 GitHub stars and 253 forks, is released under the AGPL-3.0 license, and was last updated on GitHub on 4 October 2026.

Repository
felinics/Memoh
GitHub stars
2,639
Forks
253
License
AGPL-3.0
Last updated on GitHub
4 October 2026
Category
Agents
Listing
Hidden gem β€” a shorter page for a lesser-known project
Official site
Official site
// how it works

How it works

  1. Deploy Memoh

    Self-host the platform.

  2. Add your key

    Bring your own model API key.

  3. Spawn agents

    Each gets a computer, desktop, and memory.

  4. Manage the fleet

    Monitor agents across their machines.

// features

Why people love it

Per-agent computers

An isolated machine per agent.

Long-term memory

Agents remember across sessions.

Bring your own key

Use your own model keys.

Coding agents

Host Claude Code, Codex, and more.

Networked desktops

Full desktop environments per agent.

Fleet management

Run many agents at once.

Advertisement
// run it

Run it yourself

  1. Step 1 β€” Clone the repo

    Download the source and move into the project folder.

    git clone https://github.com/felinics/Memoh.git
    cd Memoh
  2. Step 2 β€” Check the README

    Every project documents its own setup; start with install requirements.

    cat README.md | head -60
  3. Step 3 β€” Run it

    Follow the project’s launch command β€” Docker, pip, npm, or a binary release.

    # see README for the exact launch command
// use cases

What you can build

Advertisement
// faq

Questions, answered

Who maintains Memoh?

It is an open-source project at felinics/Memoh with 253 forks.

Is Memoh free for commercial use?

It is AGPL-3.0, which has strong copyleft terms; get legal advice first.

What hardware does each agent need?

Each agent gets a virtual computer, so plan capacity accordingly.

Which models work?

You bring your own key, so any supported model API.

How is this different from Docker?

It is a full managed platform: desktops, networking, and memory per agent.

// keep exploring

Similar tools

AI Agents

nanobot

Ultra-lightweight, open-source, self-hosted personal AI agent framework in Python with WebUI, tools, memory, MCP, multi-agent workflows, automation, and chat…

β˜… 48.8k starsMIT
View guide β†’
AI Agents

agent-zero

Agent Zero AI framework

β˜… 19.4k starsUnknown
View guide β†’
AI Agents

Openai Agents Python

A lightweight, powerful framework for multi-agent workflows

β˜… 29.8k starsMIT
View guide β†’
AI Agents

agent-memory

Long-term memory runtime for AI agents β€” plain Markdown as truth, local ranked retrieval, no API key

β˜… 2.6k starsMIT
View guide β†’
AI Agents

OpenMausBot

OpenMausBot β€” the open-source Grok Bot alternative: your own team of AI bots in a chat app, each a real local agent with its own personality, model and cloud computer

β˜… 4.9k starsApache-2.0
View guide β†’
AI Agents

memobase

User-profile-based long-term memory for AI chatbot applications

β˜… 2.9k starsApache-2.0
View guide β†’
// alternatives

Alternatives in Agents

The most-starred other Agents projects in the LibreAI catalogue β€” the realistic shortlist if Memoh is not the right fit.

Get Memoh

Free and open β€” grab the source, star the repo, and start creating.