Starter
The always-on gateway without browser automation.
$10/mo
- + 500m CPU request / 1000m limit
- + 1536Mi memory request / 2048Mi limit
- + 5Gi persistent RWO storage
- + No Chromium browser
OpenClaw cloud // isolated runtime
Managed OpenClaw VPS hosting, ready for your agent.
Run a personal AI agent in its own managed container from $10/month. Nova handles the pinned runtime, HTTPS route, resource limits, persistent disk, and updates; you bring the model key.
// PLANS
Two measured tiers, one isolated agent
OpenClaw is open source, but model calls are not included. Bring your own provider key and pay that provider directly for tokens.
Pro
More headroom plus bundled Chromium for browser-capable workflows.
$20/mo
- + 1000m CPU request / 2000m limit
- + 3072Mi memory request / 4096Mi limit
- + 15Gi persistent RWO storage
- + Chromium included
Measured basis: idle anonymous memory was about 895–913 MiB, gateway RSS about 877 MB, and peak cgroup use about 1.26 GB. The gateway was OOM-killed under a 1 GiB limit. Starter requests 1.5 GiB, which costs about $9/month at the planning estimate of $6/GiB-month—roughly break-even on memory before CPU, disk, egress, payment fees, support, and shared overhead.
// MANAGED, NOT RAW ROOT
The VPS resources you searched for, without server chores
A raw VPS gives you SSH and root, then leaves patching, container hardening, routing, process recovery, and upgrades to you. Nova’s managed OpenClaw hosting gives you the Control UI and focused lifecycle controls instead—there is no SSH or root access.
What Nova operates
- + Pinned OpenClaw runtime and approved version updates
- + Per-instance container, persistent disk, limits, and HTTPS hostname
- + Edge credentials plus OpenClaw gateway-token authentication
- + Dashboard setup for provider keys and Telegram, Discord, or Slack
// SECURITY BOUNDARY
Isolation with an honest internet boundary
Contained by default
Each agent runs non-root in its own restricted container and namespace, with a read-only root filesystem, fixed resources, and no access to cluster-private or Nova platform services.
Public HTTPS remains open
The agent needs outbound public HTTPS for model providers, channels, and packages. It can therefore send data or injected secrets to non-Nova public HTTPS endpoints. Isolation limits reach; it does not make an agent incapable of exfiltration.
There are no Nova-provided backups or point-in-time restore for the OpenClaw volume. Disk loss, application corruption, cancellation purge, or customer deletion can permanently remove state.
// SETUP
From checkout to your first channel
- 01
Choose a tier
Sign in, complete checkout, and Nova provisions a standalone instance—no WordPress site or subdomain choice required.
- 02
Add your model key
Supply your own supported provider key in the dashboard. Nova does not bundle model tokens.
- 03
Open the Control UI
Reveal or rotate the edge credentials, reveal the gateway token, then use both authentication prompts.
- 04
Connect a channel
Add Telegram, Discord, or Slack credentials and wait for the runtime to confirm activation.
Mac Mini, Raspberry Pi, Windows, or OpenClaw cloud?
A Mac Mini offers strong local hardware but must stay awake and reachable. A Raspberry Pi can be inexpensive but needs careful memory, cooling, storage, and architecture management. Windows can host a suitable Linux or container environment, but the agent stops when that machine sleeps. Nova trades home-hardware control for an always-on managed container with fixed resources and no root access.
// FAQ
OpenClaw hosting questions
What are the OpenClaw system requirements?
OpenClaw runs an always-on Node gateway and keeps its state on persistent storage. Nova Starter supplies 500m requested CPU, a 1000m CPU limit, 1536Mi requested memory, a 2048Mi memory limit, and 5Gi of storage. Pro supplies 1000m/2000m CPU, 3072Mi/4096Mi memory, 15Gi of storage, and Chromium.
How much RAM does OpenClaw need?
Nova measured about 895–913 MiB of anonymous memory at idle, about 877 MB gateway RSS, and roughly 1.26 GB cgroup peak. A 1 GiB limit OOM-killed the gateway; the final Starter limit is therefore 2048Mi, with a 1536Mi request and a Node heap cap to bound growth.
How much does OpenClaw cost, and is OpenClaw free?
OpenClaw itself is open source. Nova managed hosting costs $10/month for Starter or $20/month for Pro. Model usage is separate: you bring your own supported provider key and pay that provider directly.
How does hosted OpenClaw compare with a Mac Mini?
A Mac Mini can be a capable home host if you already own one and keep it awake, powered, patched, reachable, and secured. Nova is the simpler always-on option when you do not want to operate home hardware or expose it to the internet.
Can OpenClaw run on a Raspberry Pi?
It can run on compatible Raspberry Pi hardware, but memory, storage durability, cooling, and architecture support require hands-on care. Nova provides fixed cloud resources and avoids maintaining a home device.
Can OpenClaw run on Windows?
You can operate OpenClaw from Windows with an appropriate local container or Linux environment. Hosted OpenClaw moves the Linux runtime off your PC, so it keeps running when the PC sleeps or disconnects.
How do I connect Claude or another model provider?
After checkout, add your own Anthropic, OpenAI, xAI, Gemini, Groq, DeepSeek, or OpenRouter key in the Nova dashboard. The key is stored as an instance secret and applied when the managed container restarts.
Are model tokens included?
No. Nova does not resell model tokens. Hosting and model usage are separate charges, and usage is billed by the provider whose key you connect.
Do I get SSH or root access?
No. This is a managed, isolated container rather than a raw root VPS. You use OpenClaw’s Control UI plus Nova controls for credentials, connections, restart, approved updates, logs, and usage.
Can I cancel OpenClaw hosting?
Yes. You can schedule cancellation from your account. When the subscription ends, the instance enters a seven-day retention period before its persistent data is permanently purged.