OpenHuman Docker Deployment Guide — One-Click Setup with Docker Compose

Deploying OpenHuman with Docker is the fastest way to set up a development or testing environment. This guide walks you through the entire process.

Prerequisites

• Docker 24+ and Docker Compose 2.20+ installed
• At least 4GB RAM allocated to Docker
• Git (to clone the repository)

Quick Start with Docker Compose

Create a docker-compose.yml file:

version: "3.8"
services:
  openhuman:
    image: ghcr.io/tinyhumansai/openhuman:latest
    ports:
      - "3000:3000"
    volumes:
      - ./openhuman-data:/home/openhuman/.openhuman
    environment:
      - OPENAI_API_KEY=your-api-key-here
    restart: unless-stopped

Then run:

docker compose up -d

Accessing the Container

Once running, OpenHuman is available at http://localhost:3000. The data directory is persisted at ./openhuman-data/ on your host machine.

Configuration

You can mount a custom config file:

volumes:
  - ./config.toml:/home/openhuman/.openhuman/config.toml
  - ./openhuman-data:/home/openhuman/.openhuman

Using Local Models with Docker

To use Ollama alongside OpenHuman in Docker:

version: "3.8"
services:
  openhuman:
    image: ghcr.io/tinyhumansai/openhuman:latest
    ports:
      - "3000:3000"
    volumes:
      - ./openhuman-data:/home/openhuman/.openhuman
    depends_on:
      - ollama
  ollama:
    image: ollama/ollama:latest
    ports:
      - "11434:11434"
    volumes:
      - ./ollama-data:/root/.ollama

Updating the Container

docker compose pull docker compose up -d