8 Best Hermes Agent Hosting Providers in 2026, Ranked on Backups, Updates and Verified Price

Hermes Agent from Nous Research is MIT-licensed, free, and runs on hardware you choose. That makes hosting a question about state rather than raw CPU: the skills it writes for itself, its recall index and its configuration all sit on the disk you picked, with no vendor-side copy unless you arranged one. Ranked on state protection, who is responsible for updates, and what the price actually is, Cloudways Managed AI Agents ($9.99–$79.99/mo) comes first for automated daily backups and validated agent updates. DigitalOcean’s Marketplace 1-Click App is second. Kamatera ($4/mo, 99.95% uptime guarantee) is third for anyone who wants a published uptime figure and per-minute billing.

The reason this ordering differs from a normal VPS round-up is Hermes’ learning loop. A background process takes over once a task wraps with five or more tool calls, condensing the trajectory into a reusable Markdown skill file that carries a YAML frontmatter header. The agent also improves those skills as it uses them, recalls past conversations through full-text search, and builds a persistent model of you. A server that has run Hermes for six months holds something you cannot re-download. Lose it and you lose the accumulated procedure, then pay tokens to re-derive it.

Everything below is a comparison of published terms, specifications and prices as read on 28–29 August 2026. Where a vendor publishes no uptime figure, this guide says so instead of guessing.

Key takeaways

• Cloudways is the only option here that includes automated daily backups and vendor-validated agent version updates in the plan, from $9.99/mo.
• Only two published uptime guarantees exist across these eight: Contabo at 99.9% and Kamatera at 99.95%. Nobody else publishes one on the relevant page.
• Hosting lands between $5 and $20 a month almost everywhere; model tokens are the dominant line item, not the server.
• Hermes v0.20.0 arrived on 3 August 2026 carrying roughly 3,650 commits since v0.19.0 — “who applies updates” is a real scheduling decision.
• Ecosystem-wide, researchers counted 1,184 confirmed malicious agent skills across 10,700+ packages by 1 March 2026. Not Hermes incidents, but the reason skills deserve review before installation.

Hermes Agent hosting at a glance

How we ranked these

Three weights, in this order.

State protection (heaviest). Does the plan include automated backups with a restore path, or is protecting the skills directory, the memory and recall database, and the configuration entirely your job? An agent that rewrites its own capability set is a stateful workload, and snapshots matter more than clock speed.

#1 Best Overall

Sale

Elebase USB to USB C Adapter for iPhone 17 4Pack,USBC Car Charger Adapter

• Read Before You Buy — No Video Output: These adapters support charging and USB 2.0 data transfer, but cannot transmit video signals. Except for standard USB webcams (which use USB data only), they are not compatible with HDMI/DisplayPort cables, video-capable USB-C hubs, or docking stations with video output.
• Convert USB-A Ports to USB-C: Designed to connect USB-C earphones, cables, flash drives, card readers, and other USB-C accessories to standard USB-A ports. Plug-and-play with no drivers or software required.
• Aluminum Alloy Housing: Built with a sturdy aluminum alloy shell that aids in heat dissipation and protects against daily wear and scratches. Designed to maintain a stable and secure connection.
• Compact & Travel-Friendly: The ultra-compact design allows the adapter to stay plugged into your device without blocking adjacent ports or adding bulk, reducing wear and tear on your original USB ports.
• 12-Month Warranty: Backed by a 12-month manufacturer warranty for peace of mind. Designed to meet strict quality control standards for reliable everyday performance.

Update responsibility. Roughly 3,650 commits, 1,400 merged pull requests and 1,200 closed issues separate v0.19.0 from v0.20.0. Somebody has to decide when that lands on your box. A host that validates and applies agent versions is doing work you would otherwise schedule yourself.

Verified price. Published, dated figures only. Where a vendor’s own materials disagree with themselves, or where pricing came from a third-party tracker, the entry says so and ranks accordingly.

Plain admission: we did not benchmark these hosts. No synthetic runs, no latency tables, no disk throughput figures. This is a comparison of published terms, defaults and prices read on 28–29 August 2026, plus judgement about how those defaults suit an agent that accumulates state. Confirm current pricing before you buy — vendors change it.

1. Cloudways Managed AI Agents: best for protected state and validated updates

Cloudways, part of DigitalOcean, sells a managed home for autonomous agents and supports both Hermes and OpenClaw. The company describes Hermes as “the agent built for safe, isolated multi-agent work”. You get a dedicated instance per customer rather than shared tenancy.

Plans run Scout at $9.99/mo (1 vCPU, 2 GB RAM, 50 GB storage, 2 TB transfer), Operator at $19.99 (2 vCPU, 4 GB, 80 GB, 3 TB), Squad at $39.99 (4 vCPU, 8 GB, 160 GB, 5 TB) and Swarm at $79.99 (8 vCPU, 16 GB, 320 GB, 6 TB). A 50% limited-time discount was running at the time of writing. The plan bundles Cloudways-validated agent version updates, automated daily backups with restore, managed SSL, SSH access, region selection and 24/7 support. LLM costs stay with your provider — the company states that explicitly.

It tops this list for a structural reason rather than a marketing one. Two of the three things that make Hermes awkward to host — protecting state that only exists on your disk, and deciding when a very fast-moving codebase updates — are the two things this plan covers by default. Daily backups plus validated updates map directly onto the failure modes of a self-modifying agent.

Suits anyone who wants the agent’s memory to survive a mistake without building the backup job themselves, and teams who would rather not own a release calendar.

Watch for: the product page lists no uptime percentage or SLA. Update timing is Cloudways’ call, not yours, which is the trade you are making — if you need a specific Hermes version pinned on a specific day, a plain VPS gives you that control back.

2. DigitalOcean: best for a hardened one-click start you still own

DigitalOcean publishes a Hermes Agent Marketplace 1-Click App alongside its OpenClaw image, plus explainer content covering what Hermes is and how self-hosting works. You deploy a Droplet, the image handles the install, and the box is yours from there.

What earns second place is the documented house style for agent images. On the OpenClaw image, DigitalOcean publishes its hardening approach in the open: a Caddy reverse proxy with Let’s Encrypt TLS issued against the IP, a generated gateway token, device pairing, ephemeral Docker containers for execution, fail2ban, unattended upgrades and rate-limited ports. That is a meaningful signal about defaults — unattended upgrades keep the operating system patched without you, and ephemeral execution containers shrink the blast radius when a tool call goes wrong. Confirm which of those specifics ship in the Hermes image when you deploy it.

Suits developers who want a working install in minutes and full root control afterwards, with a large documentation base behind the platform.

Rank #2

Sale

Anker USB-C Hub, 5-in-1 USB Hub for Laptops, 4K HDMI Multiport Adapter

• 5-in-1 USB-C Hub: Experience comprehensive connectivity featuring a Power Delivery input, two USB-A 2.0 ports, a USB-A 3.0 port, and an HDMI port. (Note: The USB-C power delivery input port is only for connecting an external wall charger to power your laptop and cannot power peripheral devices.)
• 90W Pass-Through Charging: Achieve optimal charging with 90W pass-through power to your laptop, supported by a total input of 100W, with the hub reserving 10W for operational efficiency. (Note: Wall charger not included.)
• Quick Data Transfers: Accelerate your productivity with rapid data transfers using a high-speed 5Gbps USB 3.0 port and two 480Mbps USB 2.0 ports.
• 4K HDMI Display: Enhance your visual experience with a hub capable of delivering 4K resolution at 30Hz in both mirror and extend modes. Please note that this hub is compatible with MacBook (macOS 12 and newer), Windows 10 and 11, ChromeOS, and laptops equipped with DP Alt Mode and Power Delivery. Note: This device is not compatible with Linux.
• What You Get: Anker USB-C Hub (5-in-1, 4K HDMI), welcome guide, 18-month warranty, and our friendly customer service.

Watch for: the price. DigitalOcean’s own OpenClaw materials cite both $12/month and a $24/month minimum in different places, so check the current Droplet size recommendation and cost on the live listing before you commit. Marketplace listings carry no SLA figure. And nothing in the listing promises backups of your agent’s state — assume scheduling snapshots is your job, and verify what your Droplet actually includes.

3. Kamatera: best for a published uptime figure and per-minute billing

Kamatera sells configurable cloud servers with an entry instance at $4/month: 1 vCPU at 2667 MHz, 1 GB RAM, 20 GB NVMe SSD, 5 TB traffic, Ubuntu 24.04. Hourly servers get billed by the minute, with $0.014/hour as one listed example. There are 18 global locations, a 30-day trial worth up to $100, and no long-term commitment.

It publishes a 99.95% uptime guarantee covering servers, CPUs, firewalls, load balancers and storage — the highest published figure anywhere in this comparison, and one of only two that exist at all.

The per-minute billing matters more for Hermes than it looks. The project publishes no fixed RAM or CPU requirement; the honest guidance is to monitor real demand and resize from evidence. A provider that bills by the minute lets you test a 2 GB box against a 4 GB box on the same workload for the price of a coffee, then keep whichever the browser automation actually needs.

Suits operators who want a contractual uptime number and the freedom to resize without a monthly penalty.

Watch for: no Hermes template, so you are running the single-command installer yourself, and 1 GB of RAM gets thin quickly once a headless browser or Docker-isolated tooling is running. Budget for the next size up rather than the headline $4.

4. Contabo: best value when memory is the binding constraint

Contabo’s Cloud VPS 4 is 4 vCPU, 8 GB RAM and 100 GB SSD for €5.50/month on a 24-month term, tax included. VPS 6 (6 vCPU, 12 GB) runs €7.50, and VPS 8 (8 vCPU, 24 GB) runs €14.00. Traffic is unlimited, DDoS protection is free, there are 11 locations, and the company publishes a 99.9% uptime guarantee.

Eight gigabytes for the price other vendors charge for two is the whole argument. Memory is what actually binds a Hermes instance: the moment a headless browser opens, or several Docker-isolated tools run in parallel, RAM is the ceiling you hit first. If your plan involves browser automation plus sub-agent delegation plus scheduled jobs, Contabo puts you on the third rung of the sizing ladder for entry-level money.

Suits self-hosters comfortable with a bare Linux box who want headroom rather than hand-holding.

Watch for: that headline price is a 24-month commitment, which sits awkwardly against advice to resize from evidence. The vCPUs are shared. Backup is a paid add-on, not a default — on a list ranked by state protection, that is what keeps Contabo out of the top three, because the one thing you cannot re-create is exactly the thing you must pay extra to protect. There are no agent-specific docs and no compliance story.

5. Hetzner Cloud: best cheapest competent always-on box

No Hermes template here, which matters less than it sounds — the project’s installer is a single command on Linux, macOS, WSL2 and Android via Termux, and it pulls in prerequisites automatically. CX22 is €4.49/month and CAX11 is €5.99/month, plus €0.50 for an IPv4 address, all excluding VAT. Locations cover Germany, Finland, Singapore and two sites in the United States.

Rank #3

Anker USB C Hub, 7in1 Multi-Port USB Adapter, 4K@60Hz USBC to HDMI Splitter

• Sleek 7-in-1 USB-C Hub: Features an HDMI port, two USB-A 3.0 ports, and a USB-C data port, each providing 5Gbps transfer speeds. It also includes a USB-C PD input port for charging up to 100W and dual SD and TF card slots, all in a compact design.
• Flawless 4K@60Hz Video with HDMI: Delivers exceptional clarity and smoothness with its 4K@60Hz HDMI port, making it ideal for high-definition presentations and entertainment. (Note: Only the HDMI port supports video projection; the USB-C port is for data transfer only.)
• Double Up on Efficiency: The two USB-A 3.0 ports and a USB-C port support a fast 5Gbps data rate, significantly boosting your transfer speeds and improving productivity.
• Fast and Reliable 85W Charging: Offers high-capacity, speedy charging for laptops up to 85W, so you spend less time tethered to an outlet and more time being productive.
• What You Get: Anker USB-C Hub (7-in-1), welcome guide, 18-month warranty, and our friendly customer service.

Hetzner’s reputation is built on price-to-performance and a clean control panel, and for a Hermes instance that mostly idles — chat surfaces, light tools, an external model provider — this is the cheapest way to keep a persistent agent awake around the clock. The documentation’s own framing is that Hermes “lives wherever you put it — a $5 VPS, a GPU cluster, or serverless infrastructure”, and this is squarely the $5 VPS.

Suits people who genuinely enjoy operating their own server and want the lowest recurring bill for an always-on agent.

Watch for: everything is yours. Patching, firewalling, snapshot scheduling and execution isolation are all setup you perform and maintain, including the backup job that protects the skills directory and recall database. Hetzner publishes no uptime figure for its cloud plans. Add the IPv4 charge and VAT before comparing against the dollar prices above.

6. Vultr: best for putting the agent close to you

Vultr operates 32 data centres across 19 countries with hourly billing. Cloud Compute Regular is listed at about $5/month (1 vCPU, 1 GB RAM, 25 GB SSD, 1 TB transfer) and about $10/month (1 vCPU, 2 GB, 55 GB SSD, 2 TB), with High Performance instances listed from about $6/month.

Sourcing note, because it changes how much weight to put on those numbers: these figures come from third-party pricing trackers, not from Vultr’s own page, which was unreadable on the day of checking (29 August 2026). Treat them as approximate and confirm on the vendor’s site before you buy.

The genuine draw is geography. Hermes talks to more than 20 messaging surfaces and drives browser automation against real websites; putting the instance in the same region as the services it touches and the people it answers takes latency out of every interaction. Thirty-two locations is the widest choice on this list after Kamatera’s 18.

Suits anyone whose priority is regional placement and hourly, commitment-free billing.

Watch for: the unverified pricing above, no agent template, and no uptime figure we could verify. As with Hetzner and Contabo, backups and updates are entirely your responsibility — which, on this list’s ranking axis, is why a wide region map alone cannot lift it higher.

7. Modal or Daytona: best for bursting heavy execution off a small server

These two are not hosts for the agent. They are execution backends — two of the six Hermes supports, alongside local, Docker, SSH and Singularity. Their distinguishing property is serverless persistence: the environment hibernates when idle and costs nearly nothing between tasks.

This serverless persistence enables the most underrated architecture in Hermes hosting. A small, cheap, always-on server holds the agent, its skills and its memory. Heavy or risky tool execution bursts out to an isolated sandbox that wakes on demand and sleeps afterwards. You pay a flat few pounds a month for the part that must never go away, and metered pennies for the part that runs occasionally. The blast radius of a bad command also stops at the sandbox rather than the machine holding your accumulated state.

Suits established instances with a known, bursty workload — long scraping runs, batch code execution, anything you would rather not run beside your recall database.

Rank #4

UGREEN USB to USB C Adapter Combo 4-Pack, 10Gbps Type C Converter

• Dual Converters, Infinite Potential：Includes 2× USB C male to USB A female adapters and 2× USB A male to USB C female adapters. Perfect for a wide range of uses—tablets with Bluetooth keyboards, expand USB ports on macbook, and more. Two different converters for all your daily needs
• Next-Level 10Gbps & 3A Charging: No more slow 480Mbps, this usb to usb c adapter has a transfer speed of up to 10Gbps, allowing you to do more transferring in less time. This usb adapter fits both USB A and USB C charger, supporting up to 3A fast charging
• Upgraded Exquisite Craftsmanship: With an aluminum alloy housing and metal connector, the usbc to usb adapter is extremely durable and sturdy. Rigorously tested to withstand more than 10,000 times of plugging and unplugging, ensuring long-lasting performance
• Broad Compatible: The usb c to usb adapter widely supports all USB C/ USB A devices like laptops, tablets, cellphones, car chargers, and phone chargers. Such as compatible with MacBook Pro/Air 2023/2022, Thunderbolt 4/3 Devices,Apple MagSafe Watch 9/8/7/SE/Ultra, iPad Pro 2022/2021, Samsung Galaxy S23/S20/S10, and iPhone 17/16/15 Pro. Plug and play
• Please Note: To reach 10Gbps speed, keep the cable under 3.3 ft. For USB A Male to USB C adapters, try flipping the USB C connector. USB C Male to USB A adapters support bidirectional 10Gbps transfer within 3.3 ft

Watch for: you still need to host the agent somewhere, so this is an addition to one of the options above, not a replacement. Cold starts add latency to the first call. Per-second billing is harder to forecast than a flat monthly fee. It also adds another vendor inside your trust boundary. This is an optimisation to reach for once you understand your workload, not a starting configuration. No pricing figures are quoted here because none were verified.

8. Own hardware and the desktop apps: best for one user with nothing sensitive

Nous Research ships Windows and macOS desktop installers, and the shell installer covers Linux, macOS, WSL2 and Android via Termux. Native Windows also has a PowerShell installer, though it is marked experimental and WSL2 is the recommended route. Running Hermes on a machine you already own costs nothing beyond electricity, and for a single person experimenting with the 60-plus built-in tools it is a legitimate deployment.

It ranks eighth because it is last on every axis this list weighs. State protection is the weakest of any option here: personal machines are rarely on a backup schedule that could actually restore a six-month skill library. Updates are entirely manual. And there is no price to verify because there is no service.

Suits evaluation, learning the tool, and workloads with no schedule attached.

Watch for: scheduled and background work stops when the machine sleeps, so cron tasks silently do not happen. A power cut or an ISP outage ends jobs mid-flight. Most importantly, an agent holding terminal access, file editing and code execution sits alongside everything else on that machine — your documents, your keys, your browser profile. That is a different risk posture from a $5 server that contains nothing but the agent.

How to choose

Start from workload, not from price. The project publishes no fixed RAM or CPU requirement, so treat the ladder below as guidance rather than a specification, and resize once you have watched real demand for a fortnight.

Then answer three questions honestly. Who restores this if the disk dies? If the answer is “nobody has written that job yet”, pay for a plan that includes backups. Who applies updates? If you will not schedule it, choose a host that validates versions for you. Is the price the price? Add VAT, IPv4 charges and contract length before comparing.

Two options considered and left off the numbered list. GPU hosts — RunPod, Vast.ai and Hyperstack are the names in this market — matter only if you are running models locally through Ollama or vLLM, and they bill per GPU-hour; no rates are quoted here because none were verified. A small CPU server combined with a metered API key will, for most workloads, cost far less than a rented GPU sitting idle between tasks. Akamai Cloud, formerly Linode, runs 24 or more data centres across North America (10), Europe (7), Asia-Pacific (7), plus São Paulo and Jakarta, and positions itself on “no hidden fees, no egress surprises, and no vendor lock-in”. Its prices were not readable on the day, so no figure appears here and none should be assumed.

What a month actually costs

One cost argument is specific to Hermes. Because the learning loop turns repeated work into stored skills, a well-used instance tends to spend fewer tokens on the same task over time — the procedure is recalled rather than re-derived. That makes the case for hosting on a machine that never gets wiped partly a financial one, not just an operational one.

Frequently asked questions

Do I need a host with a Hermes template?

No. The official installer is a single command on Linux, macOS, WSL2 and Android via Termux, and it installs prerequisites automatically. A template saves perhaps twenty minutes on first deployment. What a template can signal is that the vendor has thought about agent-specific defaults — DigitalOcean documents Caddy with TLS, a generated gateway token, device pairing, ephemeral execution containers, fail2ban and rate-limited ports on its OpenClaw image, which is more useful than the time saved. Hetzner, Contabo, Kamatera and Vultr have no Hermes template, and that alone should not rule them out.

How much RAM does Hermes Agent need?

The project publishes no minimum, and anyone quoting one is guessing. The practical pattern is that memory binds first, and it binds the moment a headless browser opens. A chat-only instance calling an external model provider is comfortable on 2 GB and idles most of the day. Add browser automation, scheduled jobs and Docker-isolated execution and 4 GB is the sensible floor. Several concurrent projects with sub-agent delegation want 8 GB, and by that point disk becomes the constraint as skills, logs, the recall index and downloads accumulate. Monitor demand for a fortnight, then resize from evidence.

Best Value

Sale

Anker USB C Hub, 5-in-1 USBC to HDMI Splitter with 4K Display

• 5-in-1 Connectivity: Equipped with a 4K HDMI port, a 5 Gbps USB-C data port, two 5 Gbps USB-A ports, and a USB C 100W PD-IN port. Note: The USB C 100W PD-IN port supports only charging and does not support data transfer devices such as headphones or speakers.
• Powerful Pass-Through Charging: Supports up to 85W pass-through charging so you can power up your laptop while you use the hub. Note: Pass-through charging requires a charger (not included). Note: To achieve full power for iPad, we recommend using a 45W wall charger.
• Transfer Files in Seconds: Move files to and from your laptop at speeds of up to 5 Gbps via the USB-C and USB-A data ports. Note: The USB C 5Gbps Data port does not support video output.
• HD Display: Connect to the HDMI port to stream or mirror content to an external monitor in resolutions of up to 4K@30Hz. Note: The USB-C ports do not support video output.
• What You Get: Anker 332 USB-C Hub (5-in-1), welcome guide, our worry-free 18-month warranty, and friendly customer service.

What exactly should I back up?

Three components matter here: the skills directory, the memory and recall database, and the configuration. Between them they hold everything the agent has learned that you cannot reinstall — the Markdown skill files written after tasks with five or more tool calls, the full-text recall index over past conversations, and the persistent model it has built of you. Specific file paths are not documented here because they were not verified; check the official documentation for current locations. Verify a restore actually works before you need it. A backup you have never tested is a belief, not a safeguard.

Who is responsible for updating Hermes on my server?

You are, on every option in this list except Cloudways, which ships validated agent version updates as part of the plan. The volume involved is the point: v0.20.0 landed on 3 August 2026 with roughly 3,650 commits, 1,400 merged pull requests and 1,200 closed issues since v0.19.0. Curated release notes for 0.20.0 ship with 0.21.0, so reading a changelog is not always an option at release time. Either put a recurring slot in your calendar for updates and a snapshot beforehand, or pay someone to make that call for you. The project repository is where release activity is visible.

Are community skills safe to install?

Treat a skill the way you would treat a browser extension’s permissions. Skills are portable, shareable and community-contributed through the Skills Hub, compatible with agentskills.io — genuinely useful, and genuinely an attack surface. Ecosystem-wide context, not Hermes incidents: Cisco documented a third-party agent skill performing data exfiltration and prompt injection in late January 2026, and by 1 March 2026 researchers counted 1,184 confirmed malicious skills across more than 10,700 packages. Read what a skill does before installing it, prefer running unfamiliar ones behind Docker or a serverless sandbox, and make use of MCP credential filtering.

Can I just run it on my laptop?

For evaluation and single-user work with nothing sensitive attached, yes — desktop installers exist for Windows and macOS. The failure modes are predictable. Scheduled and background tasks stop when the machine sleeps, so cron jobs quietly do not run. Power or connectivity interruptions end work mid-flight. Personal machines are rarely on a backup schedule capable of restoring a mature skill library. And an agent with terminal, file-editing and code-execution tools shares that machine with your documents, keys and browser sessions. A $5 server that contains only the agent is a cleaner boundary.

Which host gives the best uptime?

Only two of these publish a figure at all. Kamatera guarantees 99.95% across servers, CPUs, firewalls, load balancers and storage; Contabo guarantees 99.9% and adds free DDoS protection with unlimited traffic. Cloudways, DigitalOcean and Hetzner publish no uptime percentage or SLA on the relevant pages, and no verified figure exists for Vultr. That absence is not evidence of poor reliability — it means there is no contractual number to hold them to. If a written availability commitment matters to your use case, that narrows the field to two.

Verdict

Best overall, and for anyone who would not otherwise write a backup job: Cloudways Managed AI Agents, from $9.99/month. Automated daily backups with restore, plus validated agent updates, cover the two hardest parts of hosting a self-modifying agent. Accept that update timing is theirs and that no SLA is published.

Best for developers who want root and a fast start: DigitalOcean’s Hermes 1-Click App, with its documented hardening approach — but confirm the Droplet size and price on the live listing, since DigitalOcean’s own OpenClaw materials cite $12/month and a $24/month minimum in different places.

Best for a contractual uptime figure and the freedom to resize: Kamatera, $4/month entry, 99.95% guarantee, 18 locations, billed by the minute. Plan on more than the 1 GB entry instance if browser automation is in scope.

Best value when memory is the ceiling: Contabo, 4 vCPU and 8 GB for €5.50/month on 24 months with a 99.9% guarantee — provided you buy the backup add-on rather than gambling with your skills directory.

Cheapest competent always-on box: Hetzner, CX22 at €4.49/month plus €0.50 for IPv4, excluding VAT, if you are happy owning every part of the operational work.

Best once your workload is understood: a small always-on server from any of the above, paired with Modal or Daytona so heavy execution hibernates when idle and keeps risky commands away from the disk holding your agent’s memory.

Going deeper than a shortlist: One-Click or Manual? What Paying More for a Hermes Agent Server Actually Buys. Best OpenClaw Hosting in 2026: 7 Options Ranked on What Keeps a Gateway Safe and Online.

Every price, specification and uptime figure in this guide was read from the named sources on 28 and 29 August 2026. Hosting prices move, promotional discounts expire, and Vultr’s figures came from third-party trackers rather than the vendor’s own page — check the current terms before you buy.

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.

Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Support on Ko-Fi