No contracts · cancel anytime
Start at zero. Scale when you’re ready.
Plans are per agent: pick one for each agent you deploy, and deploy as many as you like. The difference is how much muscle each gets. LLM credits included on every plan.
$0/mo
Kick the tires
- AI agent
- 2 vCPU · 4 GB RAM
- 12 GB disk
- $1 LLM credits /mo
- Auto-sleeps when idle
$19/mo
Your first real sidekick
- AI agent, always on
- 2 vCPU · 4 GB RAM
- 12 GB disk
- $5 LLM credits /mo
$49/mo
For agents with a real job
- AI agent, always on
- 4 vCPU · 8 GB RAM
- 20 GB disk
- $15 LLM credits /mo
Compare plans
The whole menu, side by side.
Swipe sideways to compare plans →
| What you get | Free$0/mo | Starter$19/mo | Pro$49/mo | Max$99/mo |
|---|---|---|---|---|
| Your agent | ||||
| AI agent (OpenClaw or Hermes) | ||||
| Always on | ||||
| Auto-sleeps when idle | After 30 min | |||
| Chat app connections | Unlimited | Unlimited | Unlimited | Unlimited |
| Compute | ||||
| vCPU | 2 | 2 | 4 | 8 |
| RAM | 4 GB | 4 GB | 8 GB | 16 GB |
| Disk | 12 GB | 12 GB | 20 GB | 30 GB |
| AI usage | ||||
| LLM credits included /mo | $1 | $5 | $15 | $25 |
| Bring your own API keys | ||||
| Top up credits anytime | ||||
| Peace of mind | ||||
| Automatic updates | ||||
| Monitoring & auto-restart | ||||
| Daily backups | ||||
| Dashboard & logs | ||||
| Support | Community | Priority email | Priority email | |
Prices in USD, per agent. Deploy as many agents as you like, each on its own plan, starting free.
Billing FAQ
Money questions, straight answers.
What are LLM credits?
Your agent thinks by calling AI models, and that usage normally requires your own API keys and billing accounts. We bundle it instead: every plan includes a monthly allowance of LLM usage, so your agent works the moment it's deployed. $5 of credits covers a lot of everyday chatting; heavy research and browsing use more.
What happens if I use up my credits?
Your agent politely pauses AI responses until the next month, or you can top up credits from the dashboard on any paid plan. You can also plug in your own API keys at any time and pay your provider directly. We never silently bill you for overages.
What does auto-sleep mean on the Free plan?
After 30 minutes without activity, your free agent takes a nap to save resources. It wakes up automatically when you message it, which takes a few seconds. Paid plans stay always on, so scheduled tasks and instant replies work around the clock.
Can I run both OpenClaw and Hermes Agent?
Yes. Plans are per agent, so you can deploy as many agents as you like, each on its own plan. Many people start both on the free tier to compare them. You can also switch which agent a plan runs whenever you like.
Can I upgrade, downgrade or cancel?
Anytime, in one click from the dashboard. Upgrades apply instantly with prorated billing. Downgrades and cancellations take effect at the end of your billing month, and you can export your agent's data before you go.
Do you offer yearly billing?
Not yet. All plans are monthly with no contracts, which keeps things simple while we grow. Yearly plans with a discount are on the roadmap.