One price.
One agent.
Flat per-agent monthly pricing. Every feature included. No tiers, no add-ons, no surprises.
One subscription per agent. Billed monthly. Cancel anytime — runs through the end of the period.
- Dedicated VPS on Hetzner
- Cloudflare tunnel & public URL
- Persistent memory across sessions
- Auto-generated, versioned skills
- All gateways: Discord, Slack, WhatsApp, Telegram, Signal, email
- Bring your own model — Claude, GPT, Gemini, or local
- Dashboard, logs, and live status
The fine print,
up front.
Can I deploy multiple agents?
Yes. Each agent is its own subscription, billed independently. There's no cap on how many you run.
What happens when I cancel?
Cancellation schedules teardown at the end of the current billing period. The agent keeps running until then; we don't pro-rate or refund partial months. After teardown, the VPS and its memory are destroyed.
Where does my agent run?
On a dedicated Hetzner VPS in the region you choose at deploy time — Frankfurt, Ashburn, Helsinki, and others. Single-tenant; no shared compute.
Which models are supported?
Claude, GPT, Gemini, and self-hosted weights. You configure the model and credentials at deploy time and can change them later from the dashboard.
Is there a free trial?
Not currently. Every agent runs on real provisioned infrastructure from minute one — there's no shared sandbox to give away. The runtime itself is open-source if you'd rather self-host.