Skip to content

AI agent VPS

A VPS to run your AI agent 24/7

OpenClaw and Hermes Agent are open source AI assistants that carry out tasks for you and answer on your messaging apps. On a VPS, the agent stays reachable day and night, and its commands run on a separate machine rather than on your personal computer. The AI model itself usually runs at an API provider.

From €6.99 /month, no commitment

  • Online 24/7

    Your agent answers on Telegram, Discord or WhatsApp even when your PC is off.

  • Isolated from your personal files

    An agent that runs commands should run on its own: a dedicated VPS limits the damage in case of a mistake or a malicious instruction.

  • Few resources needed

    When the model runs at an API provider, the agent only needs a few GB of RAM; allow more if it drives a browser.

  • Hosted in France

    Dedicated vCPUs, NVMe storage and Anti-DDoS included, with no commitment.

Recommended plans

Billing period
  • Getting started

    VPS Starter

    Plenty of room for a website, an API or a few containers.

    €6.99 /moNo commitment

    Order
    • 2 dedicated vCPUs
    • 4 GB DDR4
    • 40 GB NVMe
    • 100 Mbps network
    • Full root access
    • 1 dedicated IPv4 address
    • noVNC console
    • Automatic backup
    • Firewall and reverse DNS
  • Demanding apps

    VPS Pro

    More memory for databases and resource-hungry applications.

    €18.99 /moNo commitment

    Order
    • 4 dedicated vCPUs
    • 12 GB DDR4
    • 80 GB NVMe
    • 500 Mbps network
    • Full root access
    • 1 dedicated IPv4 address
    • noVNC console
    • Automatic backup
    • Firewall and reverse DNS
  • Performance

    VPS Elite

    For running several services side by side on a 1 Gbps network.

    €25.99 /moNo commitment

    Order
    • 6 dedicated vCPUs
    • 16 GB DDR4
    • 100 GB NVMe
    • 1 Gbps network
    • Full root access
    • 1 dedicated IPv4 address
    • noVNC console
    • Automatic backup
    • Firewall and reverse DNS

Prices in euros, VAT not applicable (article 293 B of the French General Tax Code). Secure payment in the client area, automatic delivery.

See the full Linux VPS range →

How it works

  1. Order a VPS running Ubuntu LTS and create a user dedicated to the agent.
  2. Install OpenClaw or Hermes Agent with its official script, then enter your model's API key.
  3. Connect your messaging app and limit what the agent is allowed to run.

French infrastructure, verifiable numbers

  • L3/L4Anti-DDoS included

    Network attack detection and mitigation on every plan, at no extra cost.

  • 3.5 GHzXeon CPU

    Intel Xeon Gold 6138 and E5-2673 v4, up to 3.5 GHz turbo for your game servers.

  • 100%NVMe storage

    No spinning disks. Worlds and databases that respond fast.

  • < 15 msto Paris

    Infrastructure hosted in France, close to Paris for a low ping.

Hardware, protection, backups and availability in detail →

Frequently asked questions

OpenClaw or Hermes Agent: which one?

Both are open source agents installed with a script on Linux and driven from your messaging apps. Hermes Agent, published by Nous Research under the MIT licence, works with many model providers and can import an OpenClaw setup. Monthly billing with no commitment lets you try both.

Do I need a GPU for an AI agent?

No, as long as the model runs at an API provider. Our VPS have no GPU: a local model served by Ollama is possible, but limited to small models on CPU.

Is it risky to let an AI agent run commands?

An agent can run commands and read files: give it access only to what it needs. Run it as a non-root user, never expose its web interface to the Internet without authentication (use an SSH tunnel instead) and keep your API keys out of your Git repositories.

Is the agent preinstalled?

No: the VPS comes with Linux and root access, and you install the agent yourself by following our guide. You choose the version and when to update.

Other Linux VPS use cases