Deploy OpenClaw to the Cloud: One-Click Zeabur Deployment + KimiClaw Quick Start

Why Deploy to the Cloud?

While installing OpenClaw locally gives you maximum flexibility, you might run into these situations:

• 💻 Hardware limitations: OpenClaw plus models need a certain amount of computing resources
• 🌐 Need 24/7 uptime: Your Agent needs to be available around the clock
• 🚀 Don’t want to deal with setup: Python versions and dependencies are giving you headaches
• 👥 Team sharing: Multiple people need to use the same OpenClaw instance

This article will walk you through two cloud options, from simplest to most flexible.

---

Option 1: Zeabur One-Click Deployment (Recommended for Beginners)

What is Zeabur?

Zeabur is a cloud deployment platform built by a Taiwanese team, similar to Vercel / Railway, but more friendly for Chinese-speaking users. The biggest advantage is virtually zero configuration.

Deployment Steps

Step 1: Sign Up for a Zeabur Account

1. Go to zeabur.com
2. Sign in with your GitHub account (recommended) or register with Email
3. The free plan is enough for personal testing

Step 2: Create a New Project

1. Click “Create Project”
2. Select the region closest to you (recommended: Asia - Taiwan)
3. Name your project, e.g., my-openclaw

Step 3: Deploy OpenClaw

You have two options:

Option A: Deploy from Marketplace (Fastest)

1. On the project page, click “Add Service”
2. Search for OpenClaw
3. Click “Deploy”
4. Wait about 2-3 minutes for deployment to complete

Option B: Deploy from GitHub Repo (More Flexible)

1. First, fork the OpenClaw GitHub repo to your account
2. In Zeabur, click “Add Service” → “Git”
3. Select your forked repo
4. Zeabur will automatically detect the Dockerfile and deploy

Step 4: Set Up Environment Variables

After deployment, you need to configure your API Key:

1. Click on your service → “Variables”
2. Add the following environment variables:

OPENAI_API_KEY=sk-xxxxxxxxxxxxxxxx
# Or the API Key for whichever LLM you're using
GOOGLE_API_KEY=your-google-api-key

Quick reminder: Not sure how to get an API Key? Check out the AI Model API Key Guide — Google AI Studio offers a free tier, perfect for beginners.

Step 5: Get Your Access URL

1. Click “Networking” → “Generate Domain”
2. You’ll get a URL like my-openclaw-xxxx.zeabur.app
3. Open this URL — your OpenClaw is now live!

Zeabur Pricing Reference

Plan	Cost	Best for
Developer	Free	Testing, learning
Team	~$5/month	Personal production use
Pro	Pay-as-you-go	Team use

⚠️ The free plan has usage time limits. For production use, upgrading to the Team plan is recommended.

---

Option 2: KimiClaw Cloud Experience

What is KimiClaw?

KimiClaw is a cloud-based version built on the OpenClaw architecture, powered by Kimi AI (Moonshot AI). Think of it as “OpenClaw without installation.”

How Does It Compare to Self-Hosted OpenClaw?

Comparison	Self-Hosted OpenClaw	KimiClaw
Installation difficulty	Requires environment setup	Zero install, works in browser
Customization	Full control	Limited
Model selection	Any LLM	Primarily Kimi
Cost	API usage fees	Per KimiClaw plan
Offline use	✅	❌
Custom Skills	Fully customizable	Basic Skills supported

Is It Right for You?

Choose KimiClaw if you:

• Want to quickly experience OpenClaw’s concepts
• Don’t want to deal with any technical setup
• Primarily work in Chinese-language scenarios

Choose self-hosted OpenClaw if you:

• Need to connect multiple LLMs (GPT-4, Claude, etc.)
• Need fully customized Skills
• Have data security concerns

KimiClaw Quick Start

1. Go to the KimiClaw website and register an account
2. Enter the workspace — you’ll see pre-built Agent templates
3. Select a template (e.g., “Information Organizer”)
4. Type commands directly in the chat box to experience the Agent in action

---

Option 3: Docker Self-Hosted (Advanced)

If you have your own server or want maximum control:

Prerequisites

• A VPS (Linode, DigitalOcean, AWS EC2, etc.)
• Docker and Docker Compose installed
• Basic Linux command line skills

Deployment Steps

# 1. Pull the OpenClaw Docker image
docker pull openclaw/openclaw:latest

# 2. Create docker-compose.yml
cat > docker-compose.yml << 'EOF'
version: '3.8'
services:
  openclaw:
    image: openclaw/openclaw:latest
    ports:
      - "3000:3000"
    environment:
      - OPENAI_API_KEY=${OPENAI_API_KEY}
      - GOOGLE_API_KEY=${GOOGLE_API_KEY}
    volumes:
      - ./data:/app/data
    restart: unless-stopped
EOF

# 3. Create the .env file
echo "OPENAI_API_KEY=sk-your-key-here" > .env
echo "GOOGLE_API_KEY=your-google-key" >> .env

# 4. Start
docker-compose up -d

# 5. Check status
docker-compose ps

Security reminder: For production deployment, make sure to set up HTTPS (you can use Nginx + Let’s Encrypt) and firewall rules.

---

How to Choose Between the Three Options?

Do you want a zero-install experience?
├── Yes → KimiClaw (fastest to get started)
└── No → Do you have your own server?
    ├── No → Zeabur (recommended, hassle-free)
    └── Yes → Docker self-hosted (maximum flexibility)

---

Next Steps

After deployment, you can start learning OpenClaw’s core features:

• ⚙️ Set up model switching and fallback
• 🧩 Skills: Teach AI repeatable workflows
• 🤖 Agent: Your AI alter ego
• 🧠 Soul: Give your Agent memory and personality
• 📱 Connect Telegram for on-the-go access

Have questions? Head over to the Home Discussion section to chat!