Home / Servers / OpenClaw VPS

Self-host OpenClaw on your own VPS

OpenClaw is the open-source gateway that connects your messaging apps (Telegram, Discord, Slack, Signal, WhatsApp, iMessage, Matrix, Teams) to an LLM-powered agent that can actually do things: book appointments, clear inboxes, trigger deployments, manage files. Running it 24/7 on a VPS keeps the agent reachable and persistent. Servury is built for it: anonymous signup, crypto payments, full root, no logs.

OpenClaw runs comfortably on a 4 GB plan. Starting at $15.59/mo.

Why self-host OpenClaw

✓
Always reachable. Your laptop sleeps. A VPS doesn't. Your agent stays online to receive messages and execute tasks 24/7.
✓
Your messages stay yours. OpenClaw routes through your VPS, not someone else's data center. No third party sees what you're delegating to your agent.
✓
Take real action. An agent that can book, deploy, and manage files needs persistent credentials. Keep them in one place you trust.
✓
Anonymous billing. Pay with XMR or BTC. No name, no email, no card on file. The agent runs, your identity stays out of it.

Quick start: OpenClaw on Servury via Docker

Tested on Ubuntu 24. Pick a 4 GB+ plan, deploy, SSH in.

# 1. Install Docker
curl -fsSL https://get.docker.com | sh

# 2. Pull and run the official OpenClaw image with persistent storage
docker volume create openclaw_data
docker run -d --name openclaw --restart unless-stopped \
  -p 8080:8080 \
  -v openclaw_data:/data \
  -e OPENCLAW_LLM_PROVIDER=anthropic \
  -e OPENCLAW_API_KEY=sk-ant-... \
  ghcr.io/openclaw/openclaw:latest

# 3. Open http://YOUR_SERVER_IP:8080 to finish onboarding
#    Connect Telegram, Discord, Slack or Signal from the admin panel
# 4. Recommended: put Caddy or nginx in front for HTTPS

Recommended specs

Personal

Just you, 1-2 messaging platforms, light agent traffic.

2 GB RAM · 1 vCPU · 20 GB SSD
VPS-100 plan
Recommended

Multiple platforms, full chat history, fast response times.

4 GB RAM · 2 vCPU · 40 GB SSD
VPS-150 plan
Heavy use

Local LLM (Ollama) on the same box, large file workloads, multiple users.

8+ GB RAM · 4 vCPU · 80 GB SSD
VPS-200 plan or VDS

Frequently asked questions

What is OpenClaw?

OpenClaw is an open-source gateway that connects messaging apps (Telegram, Discord, Slack, Signal, WhatsApp, iMessage, Matrix, Teams) to an LLM-powered agent runtime. The agent can book appointments, manage your inbox, trigger deployments, edit files, and more, all by conversation in apps you already use.

Why run OpenClaw on a VPS instead of my laptop?

Your laptop sleeps and changes networks. A VPS stays online 24/7 with a stable IP, so your agent is always reachable and any scheduled tasks (digest emails, daily summaries) actually run.

How much VPS do I need?

2 GB RAM minimum, 4 GB recommended for real day-to-day use. If you also want to run a local LLM (Ollama) on the same box, jump to 16 GB+ or use a separate beefier server for that.

Can I use my own LLM API keys?

Yes. OpenClaw works with Anthropic Claude, OpenAI, OpenRouter, Groq, and any OpenAI-compatible endpoint including self-hosted Ollama. You set the keys in the admin panel.

Does Servury log what my OpenClaw agent does?

No. Servury runs no application-level logging on customer servers. We do not have access to anything inside your VM. With anonymous registration and crypto payment, even the billing record contains no personal info.

Will OpenClaw stay running after I close my SSH session?

Yes if you use Docker with --restart unless-stopped (shown in the quick start above), or run it under systemd. The agent keeps running across reboots and SSH disconnects.

Can I pay with crypto?

Yes. XMR, BTC, LTC, ETH, USDT, USDC and more. No KYC, no email confirmation, no name on file.

Where are servers located?

Montreal (owned hardware), New York, London, Paris, Frankfurt, Netherlands, and Singapore. Pick whichever is closest to where your agent will be reachable from for lower latency.