AiXVPScost & terms for AI / automation

Contabo Decision guide Updated 2026-11-12

Contabo vs Vultr for a Small Automation VPS

For a small automation box that runs all day in one region, I would start with Contabo Cloud VPS 4 and put the savings into backups. For a box that needs hourly billing, many locations, a clean API, or a short experiment, I would start with Vultr. The two providers are not selling the same purchase: Contabo sells a long-term bundle with a lot of RAM and disk for the price; Vultr sells flexible cloud capacity by the hour. The workload decides. A single n8n instance with a small database, a few webhooks, and an occasional media job does not need a large cloud footprint. A campaign, a multi-region webhook receiver, or a test that may disappear next week does.

Contabo review Sources
Contabo decision guide — one purchase question, one decision.
Scope. Decision draft. Contabo and Vultr product and price facts were checked on the official provider pages and the Vultr plans API on 2026-11-12. This is a planning analysis, not a hands-on benchmark, a paid order, or a provider ranking. Prices, plan availability, and terms change; verify both configurators before ordering.

What the automation box actually needs

I would write down six requirements before comparing plan cards:

  • Always-on or intermittent. A scheduled worker that must answer webhooks at 3 a.m. is an always-on purchase. A one-off migration job is an hourly purchase.
  • Region. The webhook sender and the people waiting for the result both care about network distance; one region may be enough.
  • RAM and disk. n8n itself is modest, but binary data, execution history, Docker images, and a database grow.
  • CPU. Most webhook work waits on APIs; image, video, and PDF work can make CPU the limit.
  • Port and transfer. Media jobs and large webhook payloads care about the pipe; a 200 Mbit/s port is not a 1 Gbit/s port.
  • Recovery. Snapshots, backups, and the ability to rebuild the box matter more than a benchmark.

Those requirements are the same on both providers. The difference is how each provider prices and delivers them.

The comparison that matters

The table below uses the listed plans, not a converted currency. Contabo's figures are the list and 24-month term prices, before VAT; Vultr's are monthly USD prices from its plans API. Tax, renewal, setup fees, and add-ons are separate.

ItemContabo Cloud VPS 4Vultr vc2-2c-4gbVultr vc2-4c-8gb
vCPU424
RAM8 GB4 GB8 GB
Disk100 GB SSD80 GB160 GB
Port / transfer200 Mbit/s; unlimited traffic*3 TB bandwidth4 TB bandwidth
Snapshots1 includedcheck the current plancheck the current plan
Billing24-month term price; setup fee on monthly Cloud VPS 4hourly/monthlyhourly/monthly
Locationscheck the current Contabo list3131
List / 24-month price€5.50/mo list; €4.40/mo on a 24-month term (20% off), both excl. VAT$20/mo$40/mo

*Contabo's page uses "unlimited traffic" with a port speed and fair-use terms. Read the current terms; unlimited is not a promise of unlimited port speed.

On the raw numbers, Contabo is much cheaper per RAM and disk. That is the point of its bundle. Vultr is not trying to match it on a 24-month euro price; it is selling hourly capacity in 31 locations with an API and a cloud control plane.

Where Contabo wins

RAM and disk per euro. Cloud VPS 4 gives 8 GB and 100 GB SSD at a 24-month price below Vultr's 4 GB plan, before VAT. For a single always-on n8n box with a database and a few containers, that headroom is useful.

A predictable bill within the term. If the workload is stable and the term fits, the Contabo term price is easy to budget. The catch is the term: the low figure is for the first 24 months, and the renewal or monthly price can differ. A display term price is not a renewal quote; read the renewal line separately.

A larger box for media work. Contabo's higher Core and Performance plans scale RAM and disk aggressively. A media pipeline that needs local scratch space may fit Contabo's shape better than a small Vultr plan.

Where Vultr wins

Hourly billing and a short commitment. A migration, a campaign, or a test that runs for a week is a bad fit for a 24-month term price. Vultr lets the box disappear when the job ends.

Locations. The Vultr plans API returned 31 locations for the vc2 plans used here. If the webhook source or the audience is in a region Contabo does not cover well, that can decide the purchase before price does.

API and cloud features. Vultr exposes plans and regions through an API, and its cloud features are built for scripted provisioning. If the automation creates and destroys servers, that matters.

No setup-fee surprise on a monthly plan. Contabo's Cloud VPS 4 monthly order can carry a one-time setup fee; longer terms waive it. Vultr's hourly/monthly model does not have that specific line, though it has its own add-on and bandwidth rules.

The hidden lines

The plan card is not the bill. On Contabo, I would check the term, the setup fee on a monthly Cloud VPS 4 order, VAT, the renewal price, snapshots, Auto Backup, and any extra IP or storage. On Vultr, I would check the bandwidth allowance and overage, backup and snapshot charges, IPv4/IPv6, and the hourly rate for the actual plan.

The currency is also not the same. A €4.40 24-month price, excluding VAT, and a $40 Vultr price, before tax, are not directly comparable without a tax and exchange assumption. I would compare each provider in its own checkout currency and keep the assumptions in the note.

What I would choose for three cases

An always-on single-region n8n box. Contabo Cloud VPS 4 is the first candidate. The 8 GB RAM and 100 GB disk give room for n8n, a small database, and Docker images. I would use external object storage for binary data, keep one snapshot, and add an off-server backup. If the box is media-heavy, I would compare Cloud VPS Plus 4 for the NVMe and 500 Mbit/s port.

A short campaign or a migration. Vultr vc2-2c-4gb or vc2-4c-8gb is the better shape. The job has an end date; paying for a 24-month bundle is the wrong contract.

A multi-region webhook receiver. Vultr's 31 locations and API make it the more natural starting point. Contabo can still host the central worker, but I would not force a single-region box to serve a global webhook path without measuring the latency.

What the first month looks like

A plan comparison is not a deployment plan. For the first month on either provider, I would put the same operating routine around the box:

  • Provisioning: a documented image, SSH keys, a non-root user, and a firewall. The provider's console is not the app's security boundary.
  • DNS and TLS: a hostname that resolves to the new IP, a certificate that renews, and a redirect plan for the old hostname. The setup guide covers the Contabo side; Vultr's setup page covers the Vultr side.
  • Snapshots and backups: one pre-change snapshot, one scheduled off-server copy, and one restore test. A snapshot on the same provider is not an off-site backup.
  • Monitoring: a simple uptime check, disk alert, memory alert, and a log for failed executions. The automation should fail loudly.
  • Bandwidth: a weekly read of the transfer graph. A media workflow can consume a monthly allowance faster than a text workflow.
  • Cost: the first invoice compared with the cart. A display price is not an invoice.

The first month is also when I would learn whether the box is overprovisioned. If CPU stays below the plan's headroom and disk is stable, a smaller plan may be the better renewal. If the box is already near a limit, the next plan is a measured decision.

The exit and migration costs

The cheapest plan can be the expensive choice if leaving it is hard. I would record the exit path before ordering:

  • Data transfer: how to copy the database, uploads, Docker images, and logs to a new host, and how long that takes on the current port.
  • DNS: the TTL before the move, the records to change, and the callback URLs that also need updating.
  • Contracts: the term, the notice period, and whether a refund or proration exists. A 24-month term price can make an early exit expensive.
  • Application state: the n8n encryption key, credentials, workflow exports, and the database version. A file copy alone may not rebuild the app.
  • Downtime: a maintenance window, a rollback host, and a person who can change DNS.

A provider with a low monthly price and a difficult exit is not necessarily cheaper over the life of the project. The comparison should include the cost of leaving, not only the cost of staying.

Three n8n workloads and the plan I would pick

API-only orchestration. Webhooks, JSON transforms, and API calls with little binary data. I would start with the smallest plan that gives enough RAM for n8n and a database, then watch execution history and disk. Contabo Cloud VPS 4 is the price candidate; Vultr vc2-2c-4gb is the hourly candidate.

Media pipeline. Image resize, video transcode, PDF rendering, or large uploads. The port and disk matter more than the CPU count. I would compare Contabo Cloud VPS Plus 4 (NVMe, 500 Mbit/s) with a Vultr plan that has enough disk and bandwidth, and I would keep intermediates in object storage rather than on the root disk.

Multi-region or short-lived automation. A campaign, a regional webhook receiver, or a test that may be deleted next week. Vultr's hourly model and location list are the better shape. A long-term Contabo bundle is the wrong contract for a job with an end date.

A first-year arithmetic check

Using the list and 24-month prices as a planning example, Contabo Cloud VPS 4 at €5.50/month list, excluding VAT, is €66.00 for twelve months; the 24-month term price is €4.40/month. Both figures are before setup, add-ons, VAT, and any renewal change. Vultr vc2-2c-4gb at $20/month is $240 for twelve months before tax, backups, and any bandwidth overage. The Contabo line is cheaper, but it is a different contract and a different plan shape; the Vultr line can stop when the job ends. The arithmetic is a reason to compare the term and the exit, not a reason to declare one provider cheaper for every workload.

What this does not decide

This article does not benchmark n8n throughput, webhook latency, disk IOPS, or network paths on either provider. It does not quote a Contabo setup fee, a renewal price, or a Vultr bandwidth overage. It compares the buying model and the listed plan shape.

The decision rule I would keep: choose the contract that matches the life of the job. A stable always-on box can use a long-term bundle; a temporary or multi-region job should keep the exit open. The plan card is a consequence of that choice, not a substitute for it.

Sources