exaroton Review 2026
We took exaroton apart plan by plan – live pricing, what the control panel offers for file access and backup and restore, backups, hardware and location options – then scored it on the same five criteria we apply to all 79 hosts we compare: hardware, panel, support, value and locations. Most of this review centres on exaroton panel, the panel exaroton customers actually run their servers from. This review focuses on direct file access and backup and restore, because those are the parts that usually expose a weak host fastest. exaroton is the rare host where the billing model changes how you use the server. Instead of paying a fixed monthly fee, you pay for the RAM while the server is actually running. AutoStart and AutoStop turn that into something practical rather than theoretical, which makes exaroton excellent for a private Minecraft group that plays a few evenings each week and poor value for a public server that must stay online all day.
Quick verdict
Our verdict on exaroton
Our score for exaroton is built on what the panel really lets you do, what the plans deliver for the money and how support is set up and rated by users – not on the feature list alone.
That makes the price incredibly workload-dependent. At £0.0084 per GB ($0.0111)-hour, an 8GB server costs £0.0669 ($0.0890) for every hour it is running. A group playing twelve hours a week can spend very little. The exact same server left online continuously becomes much less interesting compared with a normal monthly host.
The panel inherits much of the clarity that made Aternos popular, but exaroton removes many of the free-service restrictions. For a small modded Minecraft group, we think it is one of the easiest ways to buy exactly the amount of hosting you actually use.
exaroton at a glance
The important stuff before you buy
Score breakdown
Where the 8.4 comes from
Pricing and value
What the current price really buys
When we price exaroton, we configure the kind of server a real community would run and follow it to the final checkout total. We compare that usable configuration with the headline price rather than scoring the cheapest card.
We went through the exaroton pricing flow and checked what the configuration actually costs once the useful RAM, location and server options are selected. Pricing is brutally simple once you understand credits. One credit is worth £0.0084 ($0.0111), and server runtime costs one credit per GB of RAM per hour. Allocate 4GB and run for one hour: four credits. Allocate 16GB and run for one hour: sixteen credits. Billing is calculated more precisely than whole-hour blocks, so short sessions are not rounded up to a full hour.
The trap is comparing only the hourly number. A server that runs 24 hours a day turns 4GB into at £24.07 ($32.05) over a 30-day month before considering the small storage charge. At that point a fixed-price host can be better. exaroton wins when AutoStart and AutoStop genuinely keep runtime low.
There is also a 10GB storage limit and a small monthly storage fee of ten credits, although runtime spending offsets that charge. We treated exaroton as a usage-priced Minecraft platform, not a replacement for a large always-on public network.
| Plan / example | What you get | Current price |
|---|---|---|
| 2GB | £0.0167 ($0.0223)/hour | Usage based |
| 8GB | £0.0669 ($0.0890)/hour | Usage based |
| 16GB | £0.13 ($0.18)/hour | Usage based |
Conversion used: £1 = $1.33
Prices and product details were checked in September 2026 and can change. We compare the configured server, not just the cheapest headline card.
Control panel
exaroton panel: what it is actually like to use
For the exaroton panel, we went straight to what matters day to day: file access and backup and restore. We were looking for the normal friction points: whether common jobs are obvious, whether manual control is hidden and whether recovery tools are actually usable.
Judging file access and backup and restore, our verdict is based on the workflow rather than the dashboard styling. The exaroton panel is one of the least intimidating server-management interfaces available. If you have seen Aternos, the design language feels familiar: server software, options, players, files and backups are presented as game-server tasks rather than generic VPS administration.
The best part is how tightly the panel and billing model fit together. Changing allocated RAM is not hidden in a billing ticket. AutoStart and AutoStop are first-class options because they directly control what you spend. That makes the account feel designed around hourly hosting rather than a normal monthly product with a credit meter bolted on afterward.
For modded Minecraft, the software workflow is also much friendlier than manual hosting. We still prefer having a proper backup before changing packs or versions, but experimenting is less stressful when the panel clearly exposes the server state and the cost stops when the server stops.
Our take
What stands out about exaroton
In practice, exaroton made us think about server sessions instead of server months. We could stop treating an idle private world as something that needed to burn money overnight. That sounds obvious, but very few managed game hosts make the start-stop lifecycle this automatic.
AutoStart is the feature that turns the model from nerdy billing into something friends can actually use. Players do not need the account owner online just to press Start. They try to connect, the platform starts the server, and the group gets the convenience of on-demand hosting without a permanent monthly allocation.
The main limitation is scale. If your community expects the server to be continuously available, or if many players join unpredictably across the day, the whole reason to choose exaroton disappears. It is best for private groups, modpack seasons and occasional projects rather than a public network.
Hardware and locations
What we care about underneath the panel
We scored the hardware and location options exaroton actually attaches to its game plans instead of giving credit for the best CPU or datacentre named anywhere on the marketing site. We also held exaroton’s published DDoS and storage claims up against what each plan actually lists, rather than taking the headline marketing at face value.
This review compares exaroton’s plan and location pages with its published infrastructure claims. exaroton lets you choose between 2GB and 16GB RAM, which covers a useful range for Minecraft. The service abstracts away the underlying machine more heavily than performance-first hosts such as GHOSTCAP or DedicatedMC, so we did not choose it because of a specific CPU model.
That abstraction is part of the point. We are buying a managed Minecraft service with elastic runtime, not a named Ryzen core. For most friend-group servers, the more important question is whether the selected RAM amount handles the pack and player count without repeated restarts or memory pressure.
Storage is the area we planned around. Ten gigabytes is enough for many normal worlds, but large modpacks, maps and long-lived worlds can grow. Cloud-connected backups are therefore not just a nice extra; they are how we archived a server we are not planning to use for a while.
Network & predicted latency
What ping can you expect from exaroton?
This is a full 47-city predicted-latency model, but exaroton does not publish a fully city-resolved fixed network. The model therefore uses clearly labelled regional/country anchors where necessary. Those anchors are modelling references, not claims that the provider operates a physical facility in that exact city.
exaroton explicitly deploys dynamically within Europe or North America. These anchors represent the two selectable regions and are not fixed exaroton node locations.
Nodes / model anchors that win at least one city
Predicted RTT = ((distance km / 200) × 2) + 8 ms
UK: 10 largest cities14.4-18.7 ms
Dynamic region proxy anchors. Each player city is mapped to the geographically nearest listed node/model anchor before the propagation formula is applied.
| Player city | Nearest model anchor | Distance | Predicted ping | Class |
|---|---|---|---|---|
| London | Europe dynamic-region proxy (Frankfurt, DE) | 638 km | 14.4 ms | Excellent |
| Birmingham | Europe dynamic-region proxy (Frankfurt, DE) | 780 km | 15.8 ms | Excellent |
| Glasgow | Europe dynamic-region proxy (Frankfurt, DE) | 1,074 km | 18.7 ms | Excellent |
| Leeds | Europe dynamic-region proxy (Frankfurt, DE) | 811 km | 16.1 ms | Excellent |
| Edinburgh | Europe dynamic-region proxy (Frankfurt, DE) | 1,023 km | 18.2 ms | Excellent |
| Liverpool | Europe dynamic-region proxy (Frankfurt, DE) | 882 km | 16.8 ms | Excellent |
| Sheffield | Europe dynamic-region proxy (Frankfurt, DE) | 787 km | 15.9 ms | Excellent |
| Manchester | Europe dynamic-region proxy (Frankfurt, DE) | 838 km | 16.4 ms | Excellent |
| Bristol | Europe dynamic-region proxy (Frankfurt, DE) | 805 km | 16.1 ms | Excellent |
| Leicester | Europe dynamic-region proxy (Frankfurt, DE) | 737 km | 15.4 ms | Excellent |
US: 10 largest cities8.0-36.0 ms
Dynamic region proxy anchors. Each player city is mapped to the geographically nearest listed node/model anchor before the propagation formula is applied.
| Player city | Nearest model anchor | Distance | Predicted ping | Class |
|---|---|---|---|---|
| New York City | North America dynamic-region proxy (Chicago, IL) | 1,144 km | 19.4 ms | Excellent |
| Los Angeles | North America dynamic-region proxy (Chicago, IL) | 2,804 km | 36.0 ms | Great |
| Chicago | North America dynamic-region proxy (Chicago, IL) | 0 km | 8.0 ms | Excellent |
| Houston | North America dynamic-region proxy (Chicago, IL) | 1,516 km | 23.2 ms | Great |
| Phoenix | North America dynamic-region proxy (Chicago, IL) | 2,336 km | 31.4 ms | Great |
| Philadelphia | North America dynamic-region proxy (Chicago, IL) | 1,068 km | 18.7 ms | Excellent |
| San Antonio | North America dynamic-region proxy (Chicago, IL) | 1,694 km | 24.9 ms | Great |
| San Diego | North America dynamic-region proxy (Chicago, IL) | 2,787 km | 35.9 ms | Great |
| Dallas | North America dynamic-region proxy (Chicago, IL) | 1,295 km | 20.9 ms | Great |
| Fort Worth | North America dynamic-region proxy (Chicago, IL) | 1,327 km | 21.3 ms | Great |
EU: all 27 member-state capitals9.9-33.9 ms
Dynamic region proxy anchors. Each player city is mapped to the geographically nearest listed node/model anchor before the propagation formula is applied.
| Player city | Nearest model anchor | Distance | Predicted ping | Class |
|---|---|---|---|---|
| Vienna | Europe dynamic-region proxy (Frankfurt, DE) | 598 km | 14.0 ms | Excellent |
| Brussels | Europe dynamic-region proxy (Frankfurt, DE) | 317 km | 11.2 ms | Excellent |
| Sofia | Europe dynamic-region proxy (Frankfurt, DE) | 1,389 km | 21.9 ms | Great |
| Zagreb | Europe dynamic-region proxy (Frankfurt, DE) | 723 km | 15.2 ms | Excellent |
| Nicosia | Europe dynamic-region proxy (Frankfurt, DE) | 2,593 km | 33.9 ms | Great |
| Prague | Europe dynamic-region proxy (Frankfurt, DE) | 411 km | 12.1 ms | Excellent |
| Copenhagen | Europe dynamic-region proxy (Frankfurt, DE) | 671 km | 14.7 ms | Excellent |
| Tallinn | Europe dynamic-region proxy (Frankfurt, DE) | 1,455 km | 22.5 ms | Great |
| Helsinki | Europe dynamic-region proxy (Frankfurt, DE) | 1,515 km | 23.1 ms | Great |
| Paris | Europe dynamic-region proxy (Frankfurt, DE) | 478 km | 12.8 ms | Excellent |
| Berlin | Europe dynamic-region proxy (Frankfurt, DE) | 424 km | 12.2 ms | Excellent |
| Athens | Europe dynamic-region proxy (Frankfurt, DE) | 1,800 km | 26.0 ms | Great |
| Budapest | Europe dynamic-region proxy (Frankfurt, DE) | 811 km | 16.1 ms | Excellent |
| Dublin | Europe dynamic-region proxy (Frankfurt, DE) | 1,088 km | 18.9 ms | Excellent |
| Rome | Europe dynamic-region proxy (Frankfurt, DE) | 959 km | 17.6 ms | Excellent |
| Riga | Europe dynamic-region proxy (Frankfurt, DE) | 1,267 km | 20.7 ms | Great |
| Vilnius | Europe dynamic-region proxy (Frankfurt, DE) | 1,232 km | 20.3 ms | Great |
| Luxembourg | Europe dynamic-region proxy (Frankfurt, DE) | 191 km | 9.9 ms | Excellent |
| Valletta | Europe dynamic-region proxy (Frankfurt, DE) | 1,649 km | 24.5 ms | Great |
| Amsterdam | Europe dynamic-region proxy (Frankfurt, DE) | 363 km | 11.6 ms | Excellent |
| Warsaw | Europe dynamic-region proxy (Frankfurt, DE) | 890 km | 16.9 ms | Excellent |
| Lisbon | Europe dynamic-region proxy (Frankfurt, DE) | 1,890 km | 26.9 ms | Great |
| Bucharest | Europe dynamic-region proxy (Frankfurt, DE) | 1,454 km | 22.5 ms | Great |
| Bratislava | Europe dynamic-region proxy (Frankfurt, DE) | 650 km | 14.5 ms | Excellent |
| Ljubljana | Europe dynamic-region proxy (Frankfurt, DE) | 624 km | 14.2 ms | Excellent |
| Madrid | Europe dynamic-region proxy (Frankfurt, DE) | 1,446 km | 22.5 ms | Great |
| Stockholm | Europe dynamic-region proxy (Frankfurt, DE) | 1,187 km | 19.9 ms | Excellent |
Confidence: Low. Interpretation: Excellent <20 ms; Great 20-40 ms; Good 40-60 ms; Playable 60-100 ms; High latency 100+ ms. These are mathematical predictions, not measured ICMP or game-protocol pings. Deep research verified 2026-09-24.
Support and reputation
How exaroton support stacks up
We rate support on two things: how quickly you can realistically reach a human, based on the channels and hours exaroton publishes, and whether recent Trustpilot reviews say the answers actually move problems forward.
exaroton scores 7.8/10 for support in our breakdown above, based on the support channels it publishes and what recent Trustpilot reviews say about how problems actually get resolved.
Trustpilot was around 3.6/5 from roughly 80 reviews. Recent feedback is split in a way that fits the product: satisfied users love paying only when they play and the easy modpack workflow, while unhappy users often complain about performance, login problems or the cost when they try to run the server close to 24/7.
We did not read that as ‘bad support’ so much as a warning to buy exaroton for the use case it was designed for. If you need a permanently online server, do the monthly maths first. If you play a few nights per week, the same model can be excellent.
Pros and cons
What we liked and what held it back
- ✓ True pay-for-runtime billing
- ✓ AutoStart and AutoStop are genuinely useful
- ✓ Very easy Minecraft-focused panel
- ✓ RAM can be changed as needs change
- ✓ Good fit for occasional modded servers
- ✓ Mature documentation from the Aternos team
- ✗ Poorer value for 24/7 usage
- ✗ 10GB storage limit
- ✗ Less hardware transparency than performance-first hosts
Who should buy it?
Is exaroton the right fit?
Compare by game
See the best hosts for your game
FAQ
exaroton frequently asked questions
How does exaroton pricing work?
Is exaroton cheaper than normal Minecraft hosting?
Does exaroton support modpacks?
What are the main exaroton limits?
Who is exaroton best for?
Final verdict
Do we recommend exaroton?
We absolutely used exaroton again for the right Minecraft group. It is one of the few hosts where the pricing idea, panel and automation all reinforce each other instead of feeling like separate marketing features.
The decision is mathematical: estimate your real weekly runtime. If the server will spend most of the month stopped, exaroton can be brilliant value. If it will be online all the time, buy a conventional monthly server instead.
Affiliate disclosure: provider links may earn us a commission at no extra cost to you. Scores are based on product quality and value.