Dedicated-core KVM

Cutter

Guaranteed cores. The default choice for production.

Dedicated AMD EPYC 9004NVMe (RAID-10)From $19/mo
Short answer

Dedicated vCPU with no steal time, NVMe RAID-10, and generous transfer. This is the line most customers run web applications, databases and game servers on.

Cutter plans

Cutter 4

$ 19 /month
vCPU
2 × dedicated
RAM
4 GB
Storage
80 GB NVMe SSD
Transfer
6 TB
IPv4 / IPv6
1 / /64 routed
Configure

Cutter 8

$ 34 /month
vCPU
4 × dedicated
RAM
8 GB
Storage
160 GB NVMe SSD
Transfer
10 TB
IPv4 / IPv6
1 / /64 routed
Configure

Cutter 16

$ 62 /month
vCPU
6 × dedicated
RAM
16 GB
Storage
320 GB NVMe SSD
Transfer
15 TB
IPv4 / IPv6
1 / /64 routed
Configure

Cutter 32

$ 118 /month
vCPU
8 × dedicated
RAM
32 GB
Storage
640 GB NVMe SSD
Transfer
20 TB
IPv4 / IPv6
1 / /64 routed
Configure

Available in all 138 locations.

What Cutter is for

The Cutter line is built on dedicated AMD EPYC 9004 cores with no steal time, which means your processes get the full CPU all the time. It is the right purchase when your workload is production, sustained, or latency-sensitive, and you cannot tolerate noisy neighbours. If you are running something where a stutter costs money, this is the line. It is the default for good reason.

People run web applications, databases, game servers, CI runners, Docker hosts, and mail servers. The dedicated cores handle steady load without spiking, the NVMe RAID-10 keeps I/O predictable, and the transfer allowance suits constant traffic. The generous per-core specs keep small team deployments from outgrowing the box quickly.

Choosing a size

Choose the size by counting what you will run simultaneously, not what you might run later. A Cutter 4 runs a small web stack; a Cutter 32 handles several game servers or a busy database. If you regularly watch CPU steal time or disk I/O wait under load, you have bought too small.

Specifications

Cutter — full specification
PlanvCPURAMStorageTransferIPv4Monthly
Cutter 424 GB80 GB NVMe SSD6 TB1$19
Cutter 848 GB160 GB NVMe SSD10 TB1$34
Cutter 16616 GB320 GB NVMe SSD15 TB1$62
Cutter 32832 GB640 GB NVMe SSD20 TB1$118

Best suited to

Web applicationsDatabasesGame serversCI runnersDocker hostsMail servers

In every plan, at every tier

IncludedWhat it means
Full root / Administrator accessIncluding custom kernels, nested virtualisation and raw sockets.
True KVM virtualisationNot a container. Your own kernel, your own /proc, no shared namespace.
Custom ISO uploadBoot anything: OpenBSD, NixOS, Whonix, a hand-rolled image.
Out-of-band VNC consoleRecover a broken network config without opening a ticket.
Routed IPv6 /64Not a single address — a whole subnet, at no cost.
Always-on DDoS mitigationUp to 12 Tbps of edge scrubbing capacity, included at every tier.
Three free snapshotsInstant, and taken without pausing the instance.
Self-service reverse DNSSet PTR records from the panel; essential for mail.
Full REST API and Terraform providerEverything the panel does, scriptable.
No identity verification, everAn email address you control is the only account identifier we hold.

Frequently asked questions

Who is the Cutter line for?

Dedicated vCPU with no steal time, NVMe RAID-10, and generous transfer. This is the line most customers run web applications, databases and game servers on.

Are Cutter cores dedicated?

Yes. Each vCPU is a physical thread reserved for your instance, with no other tenant scheduled against it, so steal time is effectively zero.

Can I upgrade later?

CPU and RAM increase with a short reboot, and disk grows online. Moving down a tier requires a rebuild, so start smaller than you think you need.

What does dedicated-core mean for performance?

It means your vCPUs are mapped to physical cores exclusively. There is no time-sharing, so CPU steal time stays at zero. Sustained workloads like databases or game servers see consistent performance, and the NVMe RAID-10 keeps I/O latency low.

Can I move up a plan later?

Yes. Upgrading CPU and RAM requires a short reboot, and disk grows online. The process is self-service from the panel. Starting small is fine, but remember that moving down a tier requires a rebuild rather than a live migration.