Models Hub
Agent Tools

Memoh

A containerized AI agent platform that can connect to custom model providers.

Edit this page

Memoh is an open-source, self-hosted AI agent platform where each bot runs in its own container with persistent memory and an independent file system. It supports 9 channels — Telegram, Discord, Feishu, QQ, Matrix, WeCom, WeChat, email, and a built-in Web UI — along with rich agent capabilities such as MCP tool calling, browser automation, and scheduled tasks.

Key Features

  • Container isolation: Each bot runs in its own containerd container with a dedicated file system and network, supporting snapshots and data import/export
  • Memory engine: LLM-based fact extraction, hybrid retrieval (dense + sparse + BM25), 24-hour context loading, and memory compression and reconstruction
  • Multi-channel support: Telegram, Discord, Feishu, QQ, Matrix, WeCom, WeChat, email, Web UI
  • MCP support: Full MCP protocol support (HTTP / SSE / Stdio / OAuth), with each bot able to manage MCP connections independently
  • Browser automation: A built-in headless browser powered by Playwright, supporting web browsing, form filling, screenshots, and more
  • Web admin panel: A modern management interface built with Vue 3 + Tailwind CSS, supporting streaming conversations, tool-call visualization, file management, and more

Quick Install

Memoh is deployed with Docker. One-click install (requires Docker):

curl -fsSL https://memoh.sh | sudo sh

Or install manually:

git clone --depth 1 https://github.com/memohai/Memoh.git
cd Memoh
cp conf/app.docker.toml config.toml
# Edit the config.toml configuration file
sudo docker compose up -d

After it starts, visit http://localhost:8082. The default username and password are admin / admin123.

Connecting via NewAPI

Memoh supports connecting to any OpenAI-compatible model provider, and you can use NewAPI to manage and access various AI model services in one place.

Configuration steps

Get a NewAPI API Key

After registering and logging in to NewAPI, click "Console" in the top navigation bar, click "Token Management", then click the "Add Token" button to create a new API Key. Select the appropriate permissions, then click "Create".

Once created, click the copy button to copy the generated API Key.

copy-api-key

Add a model provider in Memoh

Log in to the Memoh web admin panel, go to the provider management page, and click NewAPI.

switch-to-newapi-provider

On the configuration page, fill in the following:

  • API Base URL: Enter the Models Hub API URL, e.g., https://modelsok.com/v1 (or http://localhost:3000/v1 for a local deployment)
  • API Key: Paste the API Key you copied from NewAPI

Click Save to complete the provider configuration.

Import models

After configuring the provider, go to the model management page and click auto-import, or manually add the models you need.

Configure a model for a bot

Go to the bot settings page, and in the model configuration section, switch the default chat model to the model you just added via the NewAPI provider, then click Save.

set_chat_model

You have now successfully configured NewAPI as Memoh's model provider. You can now chat with AI bots through Memoh's various channels (Telegram, Discord, Feishu, etc.), and all requests will be routed through NewAPI.

On this page