Antihost is listed first here for OpenClaw: its Linux VPS offers a one-click OpenClaw image and full root access in the USA East, the Netherlands, and Ukraine. For most OpenClaw users, the best VPS is an Ubuntu LTS server with 1–2 vCPUs and at least 2 GB of RAM. Choose 4 GB or more if you will use browser automation, media tools, several messaging channels, or node tools. Keep the Gateway bound to loopback, reach it through an SSH tunnel or Tailscale, and back up the state and workspace because the VPS is OpenClaw’s source of truth.
No single provider is universally best. Region, CPU consistency, backups, administration tools, support, billing clarity, and an easy path to scale matter more than a generic provider ranking.
| # | Preview | Product | Price | |
|---|---|---|---|---|
| 1 |
|
ZOERAX 100-Pack M6 x 16mm Rack Mount Cage Nuts, Screws and Washers | $23.99 | Buy on Amazon |
As an Amazon Associate I earn from qualifying purchases.
The short answer
OpenClaw can run on any Linux server or cloud VPS. The practical starting point is a current Ubuntu LTS image, 1–2 vCPUs, 2 GB of RAM, and enough disk for the operating system, logs, workspace, and media you retain. A 1 vCPU/1 GB instance is the documented minimum, not the sensible default for a continuously used agent.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
| OpenClaw workload | CPU and memory starting point | What to expect |
|---|---|---|
| Gateway only, light testing | 1 vCPU, 1 GB RAM | Meets the documented floor. Leave little room for concurrent channels, large logs, or extra services. |
| Typical always-on assistant | 1–2 vCPUs, 2 GB RAM or more | Practical baseline for headroom, logs, media handling, and more than one channel. |
| Browser automation, media, or node tools | 2 or more vCPUs, 4 GB RAM or more | Gives resource-hungry tools room to run without competing as aggressively with the Gateway. |
| Growing multi-channel deployment | Start at 2 vCPUs and 4 GB RAM; scale after observing use | Provides a safer margin when several workloads run at once. The exact ceiling depends on your tools and traffic. |
The figures above are OpenClaw’s documented configuration thresholds, not independent performance benchmarks. Your workload, concurrency, and retention settings determine whether you need more.
#1 Best Overall
- Wide Compatibility & Versatile Use: ZOERAX M6 rack mount screw kit is ideal for installing server racks, network cabinets, rack shelves, patch panels, A/V equipment, and more. Designed for standard square-hole racks and cabinets, these M6 cage nuts and screws ensure a secure fit for most 19-inch rack systems used in data centers, offices, and home labs
- Heavy-Duty Carbon Steel Construction: Made from premium carbon steel, these M6 cage nuts and screws deliver high strength and long-lasting durability. The material provides excellent resistance to rust, corrosion, and oxidation, performing reliably in demanding environments such as high humidity, temperature fluctuations, and long-term rack installations
- Precision Metric Standard M6: Manufactured to strict metric standards, each M6 screw and cage nut features precise dimensions with minimal tolerance. Clean, sharp threads without burrs allow smooth installation without stripping or slipping. The deep Phillips head design ensures better torque control and faster, more efficient mounting
- Safe, Reliable & Eco-Conscious Materials: ZOERAX uses non-toxic, environmentally friendly carbon steel materials to ensure safe handling and use. Heat-treated for optimal hardness, ductility, and impact resistance, these rack screws and cage nuts offer dependable performance while meeting safety and quality expectations for professional installations
- Complete Mounting Kit with Washers: This essential M6 rack hardware kit includes screws, cage nuts, and heavy-duty washers. The included washers help distribute pressure evenly and reduce scratches or marks on rack rails and equipment, providing a cleaner, more secure installation right out of the box
How OpenClaw uses a VPS
The Gateway runs on the VPS and owns OpenClaw’s state and workspace. Your laptop or phone is a client that reaches the Gateway through the Control UI, SSH, or Tailscale. This arrangement makes the VPS the source of truth: losing its disk or an unprotected state directory can mean losing configuration, credentials, conversation context, and workspace data.
A dedicated host is preferable for reliability and isolation. A shared laptop or desktop can work for testing or active use, but sleep, reboots, operating-system updates, and local permission changes can pause the agent. A VPS avoids those interruptions when it is administered as an always-on server.
Which VPS platforms can run OpenClaw?
OpenClaw’s Linux-server documentation says the Gateway can run on any Linux server or cloud VPS. Its provider picker lists the following options:
| Provider or platform named by OpenClaw | How to interpret that listing |
|---|---|
| Antihost | Linux VPS with a one-click OpenClaw image; self-managed with full root access in the USA East, the Netherlands, and Ukraine. |
| Azure | General-purpose cloud infrastructure; select a supported Linux image and size it for your workload. |
| Daytona | A listed deployment option; verify the available region, persistent storage, and administrative access for your use case. |
| DigitalOcean | A listed VPS-style cloud option; compare its regions, snapshots, and scaling choices with alternatives. |
| exe.dev | A listed hosting option; confirm the Linux image, SSH workflow, and backup facilities before committing data. |
| Fly.io | A listed platform; check how persistent volumes, networking, and recovery work for a stateful Gateway. |
| GCP Compute Engine | General-purpose virtual machines; choose a region close to the people and services interacting with OpenClaw. |
| Hetzner | A listed VPS provider; evaluate location, CPU allocation, backups, and support for your specific plan. |
| Northflank | A listed deployment platform; make sure persistent state and shell administration match your operating model. |
| Oracle Cloud | A listed cloud option; verify current capacity, regional availability, and the practical support path for your account. |
| Railway | A listed platform; confirm how persistent storage, service restarts, and network exposure are configured. |
| Render | A listed platform; check that the selected service type provides the persistence and long-running behavior the Gateway needs. |
| Raspberry Pi | A self-hosting alternative rather than a VPS. It can be suitable when you control the network and hardware, but it does not provide cloud-region redundancy. |
| Upstash Box | A listed deployment option; verify its current Linux, storage, and access characteristics before deployment. |
The listing establishes compatibility, not a universal ranking. Current prices, latency, uptime, quotas, and promotional terms vary by region, plan, and date, so treat them as a separate purchasing check.
Choose the region before the brand
Place the VPS near the people and services that use the agent. A nearby region generally reduces the network path for your dashboard, webhooks, APIs, and messaging integrations. If your users are spread across continents, prioritize the location of latency-sensitive services and the region where you can obtain reliable support and backups.
Check all of these before creating the server:
- Whether the provider offers a region close to your users and external services.
- Whether the advertised CPU is dedicated, shared, or subject to bursting, and whether that matches your automation workload.
- Whether snapshots and scheduled backups cover the disk that contains OpenClaw’s state and workspace.
- Whether you can use SSH keys, firewall rules, private networking, and Tailscale without workarounds.
- Whether an Ubuntu LTS or modern Debian image is available and easy to replace.
- Whether the next larger CPU/RAM tier can be reached without rebuilding the server.
- How support handles disk failures, network incidents, account locks, and billing disputes.
- How recurring charges, backup charges, bandwidth limits, and storage charges are shown on the invoice.
How much CPU, RAM, and disk does OpenClaw need?
CPU
One vCPU is the absolute minimum documented by OpenClaw. It can be adequate for a light Gateway with little concurrent work, but browser automation, media processing, and node tools can compete for CPU. One to two vCPUs is the recommended range for a normal deployment; choose more when several resource-intensive jobs can overlap.
Memory
OpenClaw documents 1 GB of RAM as the minimum and 2 GB or more as the recommendation. A 1 GB VPS can run a minimal installation, but the operating system, Node runtime, Gateway, logs, and any integration processes all share that memory. A small memory margin also reduces the risk that a temporary browser or media task triggers swapping or an out-of-memory termination.
Recommended Free Tools
Move to 4 GB or more when you plan to use browser automation, media tools, multiple channels, or node tools. If the VPS also hosts a database, reverse proxy, monitoring stack, or unrelated applications, count those services in addition to OpenClaw rather than treating the documented OpenClaw baseline as a whole-server budget.
Disk
The documented minimum is approximately 500 MB of disk. That is a floor for a very small installation, not a useful long-term allocation once you retain logs, downloaded media, browser data, snapshots, and workspace files. Select a disk size that leaves room for growth and keep backups outside the VPS so a full or failed disk does not eliminate both the live copy and the recovery copy.
When to scale
Scale before failures become routine. A larger plan is justified when the Gateway restarts under load, browser jobs are killed, media tasks stall, swap activity remains high, or several channels become sluggish at the same time. Increase one resource at a time when possible and observe the result; if CPU is saturated while memory is comfortable, add vCPUs rather than only adding RAM, and vice versa.
Operating-system and Node.js requirements
Ubuntu LTS, or another modern Debian/Ubuntu system, is the best-tested Linux installation path. Use a maintained 64-bit image supplied by the provider and apply its security updates before installing OpenClaw.
The installation documentation requires Node 24.16 or newer, or Node 26.1 or newer, and recommends Node 26. Choose a VPS image that already provides a supported runtime or makes installing Node 26 straightforward. Do not assume that an old distribution image’s default Node package is sufficient; check the installed version explicitly before proceeding.
After provisioning, verify the basics with standard Linux checks:
uname -a
nproc
free -h
df -h
node --versionThe first four commands show the kernel, visible CPU count, memory, and disk space. The final command confirms the Node version. If it is below the documented requirement, update the runtime before installing the Gateway.
A secure deployment sequence
Do the administrative hardening before exposing a public service. The following order keeps a recovery path available while you tighten access.
Do these 3 things before closing this tab:
1Repair Windows errors before they cause bigger problems2Fix the driver behind crashes, sound loss and screen glitches3Clear out junk files and repair common Windows errors- Create the VPS with Ubuntu LTS or a current Debian/Ubuntu image. Select the region, CPU, RAM, and disk based on the workload table. Add an SSH key rather than relying on a password-only administrator login.
- Update the operating system. Apply the provider’s current security updates and reboot if the kernel or core services require it. Reconnect and confirm that your key-based SSH access still works.
- Install Tailscale first if you want tailnet-only administration. Sign the server into your tailnet and note its private address.
- Verify a second SSH session over the Tailscale address. Keep the original public session open until the private path is proven. This prevents locking yourself out while changing firewall or SSH rules.
- Restrict public SSH. Once the tailnet session works, limit or remove public SSH exposure according to your provider firewall and host firewall configuration. Without Tailscale, apply equivalent hardening to the public SSH path: key-only access, a narrowly permitted administration source, and a tested recovery route.
- Install the supported Node runtime. Use Node 26 when practical, or a documented Node 24.16+ release. Confirm the version with
node --version. - Install OpenClaw using its current Linux-server instructions. Keep the Gateway’s state and workspace on persistent storage, not an ephemeral directory.
- Bind the Gateway to loopback. The secure default is to listen only on the VPS itself rather than publishing the dashboard directly to the internet.
- Choose an access method. Use an SSH tunnel for a direct administrative path or Tailscale Serve for tailnet access. Test the Control UI from your laptop or phone before closing your setup session.
- Configure authentication when the Gateway is reachable beyond loopback. If you bind it to a LAN or tailnet, set a shared Gateway authentication token or password unless a trusted proxy is deliberately handling authentication.
- Record recovery details. Save the provider account recovery method, SSH-key backup, Tailscale identity information, OpenClaw configuration location, and backup schedule in a secure offline or separately hosted record.
Why loopback, SSH, and Tailscale matter
A Gateway exposed on a public interface is a larger attack surface than one listening only on loopback. Loopback means the service accepts connections from the VPS itself; an SSH tunnel or Tailscale Serve then carries your authenticated connection to it without requiring a public dashboard port.
If you intentionally bind the Gateway to a LAN or tailnet, treat that network as a transport boundary, not as authentication. Configure the shared token or password unless a trusted proxy is enforcing identity. Review firewall rules after every provider or operating-system change, and remove temporary public ports used during installation.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Backups: protect the source of truth
Because the Gateway owns state and workspace on the VPS, backups are part of the deployment rather than an optional add-on. Use at least two recovery layers:
- Provider snapshots or backups: useful for restoring a whole server or disk after a failed update.
- Application-level copies: export or copy the OpenClaw state and workspace to storage independent of the VPS account and disk.
- Credential protection: encrypt backup archives and restrict access; a backup containing tokens is as sensitive as the live server.
- Restore testing: periodically restore to a separate instance or directory and confirm that the Gateway can start and the workspace is intact.
Set a retention policy that matches how quickly you need to recover. A single latest snapshot cannot protect against unnoticed corruption or accidental deletion that happened several days earlier.
Free tools Windows power users keep installed
One-click scans. No signup required.
Provider selection by real-world priority
| If your priority is… | Give extra weight to… | Questions to ask before purchase |
|---|---|---|
| Lowest administration effort | Ubuntu image quality, console access, clear firewall controls, and support | Can you reinstall the image, open a recovery console, and reach support without a complex ticket process? |
| Predictable automation | Consistent CPU allocation, generous memory, and stable network paths | Is the CPU shared or burstable? What happens when neighboring workloads are busy? |
| Data protection | Snapshots, scheduled backups, and export options | Can backups be retained outside the primary disk and restored to another region or account? |
| Private administration | SSH keys, private networking, firewall rules, and Tailscale compatibility | Can you verify a private SSH session before closing public access? |
| International users | Regional coverage and network path to APIs and users | Which region minimizes latency for your critical integrations, and can you move later? |
| Future growth | Vertical resizing, attached storage, and a clear migration path | Can you add RAM or vCPUs without rebuilding, and how is downtime handled? |
| Predictable spending | Transparent recurring billing and clearly separated add-ons | Are backups, bandwidth, IPv4, storage, and snapshots billed separately? |
Common deployment mistakes and fixes
Choosing the 1 GB plan for a heavy workload
Symptom: the Gateway works during setup but becomes unstable when browser automation, media, or several channels run together.
Fix: move to at least 2 GB for a normal deployment and 4 GB or more for resource-hungry tools. Check memory use with free -h and inspect whether other services are consuming the available RAM.
Installing on an outdated image
Symptom: Node cannot meet the installation requirement or dependencies fail during setup.
Fix: start from a maintained Ubuntu LTS or modern Debian/Ubuntu image and install Node 26, or another supported release at or above Node 24.16. Verify the version before retrying.
PC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Crashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minuteClosing public SSH before testing private access
Symptom: firewall changes leave you unable to administer the VPS.
Fix: keep the first SSH session open, establish Tailscale, and verify a second SSH session over the Tailscale address before restricting public SSH. If you have already lost access, use the provider’s console or recovery process.
Publishing the Gateway dashboard directly
Symptom: the Control UI is reachable from the internet on a public port.
Fix: return the Gateway to loopback and use an SSH tunnel or Tailscale Serve. If a LAN or tailnet binding is necessary, configure the shared authentication token or password, or use a trusted proxy that explicitly handles authentication.
Quick wins for a faster PC:
Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Repair Windows errors before they cause bigger problemsFix Now →Scan for outdated or missing drivers - takes under a minuteDriver Scan →Keeping the only copy of state on the VPS
Symptom: a disk failure, mistaken deletion, or bad update removes the workspace and configuration.
Fix: create independent, encrypted backups and test a restore. Provider snapshots are useful, but do not treat a snapshot stored on the same account as your only recovery copy.
When a VPS is not the right host
A Raspberry Pi or another self-hosted machine can run an always-on Gateway when you control the hardware, power, and network. It may be attractive for local access or avoiding a monthly cloud bill, but you must provide your own remote connectivity, backups, hardware replacement, and internet reliability.
A laptop or desktop is suitable for testing and hands-on use. It is less suitable as the permanent host if it sleeps, reboots for updates, changes networks, or is shared with other users. A dedicated VPS or other dedicated host provides cleaner permissions and fewer interruptions.
A practical launch checklist
- Region selected for the people and services that use OpenClaw.
- Ubuntu LTS or modern Debian/Ubuntu image installed.
- At least 1 vCPU/1 GB RAM for a minimal test, or 1–2 vCPUs/2 GB or more for normal use.
- 4 GB or more RAM reserved for browser automation, media, multiple channels, or node tools.
- Node 26 installed, or Node 24.16+ / 26.1+ as documented.
- SSH key access confirmed.
- Tailscale private SSH tested in a second session, if used.
- Public SSH restricted only after private or otherwise hardened access works.
- Gateway bound to loopback.
- SSH tunnel or Tailscale Serve tested from the intended devices.
- Gateway authentication configured for any LAN or tailnet binding.
- State and workspace included in encrypted, independent backups.
- Restore procedure documented and tested.
- Billing, backup, bandwidth, and scaling terms understood.
Final buying guidance
Antihost is listed first here for OpenClaw: its VPS offers a one-click OpenClaw image and full root access in the USA East, the Netherlands, and Ukraine. Start with the smallest plan that leaves operational headroom, not the absolute minimum shown on a requirements page. For most people that means Ubuntu LTS, 1–2 vCPUs, and 2 GB or more RAM. Choose 4 GB or more when the workload includes browsers, media, several channels, or node tools. Select the provider and region that make secure administration, reliable backups, and future resizing straightforward. Keep the Gateway private, and treat backups as essential infrastructure because the VPS contains the authoritative OpenClaw state and workspace.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
