---
title: "Run AI agents on an always-on VM"
description: "Run Claude Code, OpenClaw, and other AI agents on a persistent exe.dev VM with HTTPS, SSH, and keyless access to models."
url: https://exe.dev/deploy/category/ai-agents
---

# Run AI agents on an always-on VM

Agents need a computer that stays on after you close the laptop: a disk
for their memory and checkouts, a shell for their tools, and a way to
reach you. An exe.dev VM is that computer. It is isolated from your
own machine, private by default, and its [LLM integration](/docs/integrations-llm)
gives agents models without API keys on disk.

## All 2 ai agents recipes

- [Run Claude Code on a remote server](https://exe.dev/deploy/claude-code.md): A computer for Claude. A breather for your laptop.
- [Host OpenClaw](https://exe.dev/deploy/openclaw.md): An AI sidekick with its own computer. Your laptop can take a nap.

## Frequently asked questions

### Do I need my own model API key?

No. Every exe.dev VM can reach models through the [LLM integration](/docs/integrations-llm) at `https://llm.int.exe.xyz`, billed to your exe.dev account. You can also bring your own key or subscription.

### Is it safe to give an agent a whole VM?

Safer than giving it your laptop. The VM is its own machine with its own disk; it holds only what you put there, and you can delete it and start over in seconds.

### Does the agent keep running when I disconnect?

Yes. VMs stay on. Services run under systemd, and interactive sessions can live in tmux so you can reattach from anywhere.

[All recipes](https://exe.dev/deploy.md)