Your OpenClaw 
,
Your Server,
Your Rules.
Deploy your own AI agent server in minutes. Bring your own API key or buy credits — talk to your claw by text or voice from any device over 12+ channels.
Deploy in Clicks, Not Code
Pick a region, choose a size, hit deploy. No terminal, no SSH, no DevOps skills needed. Our guided wizard walks you through every step visually.
Hardened Servers
Every server is locked down out of the box — firewall, SSH hardening, and intrusion detection. Enable Tailnet VPN for encrypted private networking between your devices and your claw.
Native Mobile Chat
Chat with your AI agent from a real native app on iOS and Android. Voice chat, camera access, push notifications — your phone is a full AI interface.
Every Channel, One Brain
WhatsApp, Telegram, Slack, Discord, Teams, Signal, iMessage, and more. Connect any channel to your agent — no code required.
Bring Your Own Key
Use your own API key from Anthropic, OpenAI, Google, or 10+ providers — you pay the provider directly and switch models anytime. Don't want to manage keys? Buy credits and access Claude, GPT, Gemini, and more.
Always On, Always Yours
Your server runs 24/7 in the cloud. OpenClaw supports scheduled tasks, webhooks, and triggers out of the box — your claw works while you sleep.
Talk to Your Claw
Have real-time voice conversations with your AI agent. Speech-to-text and text-to-speech with natural voices — interrupt mid-response, choose from multiple voices, and talk hands-free from the mobile app.
Click Deploy
Pick a region and server size. We spin up a hardened cloud server pre-loaded with OpenClaw .
Add Your Key or Buy Credits
Paste your own API key and pay your provider directly — or buy credits to access Claude, GPT, Gemini, and more without managing keys.
Connect Channels
Link WhatsApp, Slack, Discord — whatever you use. Each channel auto-registers with the gateway.
Talk to Your Claw
Chat by text or voice from any channel or the mobile app. Your whole team can talk to your claw — each from their preferred device.
Simple, transparent pricing
One subscription covers hosting, security, and management. Bring your own API key or buy credits — your choice.
Starter
Your personal AI agent- 1 user
- All 12+ channels
- Voice chat
- BYOK or credits
- 200 MB workspace
Basic
More power, same simplicity- 1 user
- All 12+ channels
- Voice chat
- BYOK or credits
- 500 MB workspace
Pro
A shared AI co-worker for 2- 2 team members
- Role-based permissions
- All 12+ channels
- Voice chat
- 1 GB workspace
Pro+
Your team's AI co-worker — up to 10- 10 team members
- Granular permissions
- All 12+ channels
- Voice chat
- 2 GB workspace
Enterprise
Custom pricing- Custom team size
- Custom compute
- Dedicated support
- SLAs
All plans include: a dedicated server, HTTPS, firewall, SSH hardening, Tailnet VPN, voice chat, dashboard, native mobile app access, and optional automated backups. Pro and above let multiple team members talk to your OpenClaw agent — a true AI co-worker everyone on your team can collaborate with.
ZebraClaw provisions a dedicated cloud server running OpenClaw — an open-source AI agent framework. Bring your own API key, connect your channels, and talk to your claw from anywhere. No terminal commands, no config files, no DevOps experience required.
It takes four steps:
- Sign in — create an account or sign in with Google.
- Choose a plan — pick a server size that fits your workload and budget.
- Deploy — we spin up a dedicated cloud server, install OpenClaw , configure HTTPS, and harden the server automatically. This usually takes 2–3 minutes.
- Configure — add your AI provider API key, connect messaging channels, and start chatting with your agent.
Each deployment is a dedicated cloud server — not a shared container. You get:
- A running OpenClaw AI agent with your own API key (BYOK)
- HTTPS-secured gateway endpoint
- Tailnet VPN for encrypted private networking
- Built-in firewall, SSH hardening, and automatic security updates
- Dashboard for real-time chat, configuration, and monitoring
- Channel connections (WhatsApp, Slack, Discord, Telegram, and more)
- Full root SSH access if you want to customize further
You can use any major AI provider — bring your own API key. Currently supported:
- Anthropic (Claude)
- OpenAI (GPT, Codex)
- Google (Gemini)
- xAI (Grok)
- Groq, Mistral, Cerebras, OpenRouter
- GitHub Copilot
- Self-hosted — Ollama, LM Studio, and more
You can switch providers at any time from the dashboard without redeploying.
ZebraClaw gives you two ways to power your AI agent. With BYOK, you paste your own API key from any supported provider (Anthropic, OpenAI, Google, etc.) directly into your server — you pay the provider at their standard rates with full control over costs, rate limits, and models. Alternatively, you can buy credits through ZebraClaw and skip managing API keys entirely. Choose whichever fits your workflow.
Tailnet VPN (powered by Tailscale) creates an encrypted private network between your devices and your deployed server. Once enabled, your server gets a private IP accessible only to devices on your tailnet — no public exposure needed. This means you can access your server's management interfaces, SSH, and internal services through a zero-trust mesh VPN without opening additional ports to the internet.
Your API keys are sent directly to your dedicated server over an encrypted HTTPS connection. Your keys are stored on your server. You can rotate or remove keys at any time from the dashboard.
Yes. Each server is a dedicated machine — your conversations, API keys, and data never share resources with other users. Your messages go directly from your channels to your server to the AI provider you choose. We don't log, read, or store your conversations. You also get full SSH access so you can verify this yourself.
Every deployed server is hardened automatically:
- HTTPS encryption on all external connections
- Firewall configured to expose only necessary ports
- SSH key-only authentication (password login disabled)
- Intrusion detection with automatic IP banning
- Services run as unprivileged users
- Automatic security updates enabled
Absolutely. You can deploy and manage multiple servers from a single account — each with its own AI model, channels, and configuration. The dashboard gives you an overview of all your servers, and you can switch between them instantly.
You pay a simple monthly subscription based on the server size you choose. The price covers cloud hosting, provisioning, security hardening, Tailnet VPN, and the management dashboard. For AI usage, you can either bring your own API key (billed directly by your provider) or purchase credits through ZebraClaw. You can cancel anytime and your server will be cleaned up at the end of the billing period.
Yes — you get full root SSH access to your server. You can install additional software, customize OpenClaw 's configuration, set up cron jobs, or anything else you'd do on your own Linux server. It's your machine, running on your terms.
If the Control UI panel appears blank, unresponsive, or shows stale data, try these steps in order:
- Refresh the page — a simple browser refresh (F5 or Ctrl+R / Cmd+R) reloads the Control UI iframe and re-establishes the proxy session.
- Hard refresh — press Ctrl+Shift+R (Cmd+Shift+R on Mac) to bypass your browser cache.
- Switch away and back — click a different tab in the dashboard (e.g. Chat or Servers), then switch back to Control UI. This forces a fresh session.
- Check your server status — go to the Servers tab and make sure your server shows as running. If it's offline, the Control UI won't load.
If the issue persists after these steps, contact support and we'll help diagnose it.
Your claw.
Running in minutes.
No server setup. No DevOps. Bring your own key or buy credits — deploy in minutes and start talking to your claw from any device.