Best Mini PC for Proxmox (2026): Picked by What You Build

Almost every “best mini PC for Proxmox” list is organized by price: a budget pick, a mid-range pick, a high-end pick. That ordering answers a question you probably have not asked — you are not choosing a price tier, you are choosing whether a specific machine can run the thing you intend to build.

Two specifications decide that, and they are the two most guides list without explaining: how many network ports the machine has, and how many NVMe slots. Both come straight out of the Proxmox documentation, and both rule machines out.

Quick picks

• Single node, a handful of containers: Beelink EQ14. Four N150 cores and 16GB is a real ceiling, but two 2.5GbE ports mean it can still join a cluster later.
• Single node doing actual work: Beelink SER8. Eight cores, the lowest measured idle draw here, and the largest memory ceiling on paper. One network port, which is the catch.
• You want a ZFS mirror, or a cluster: Minisforum MS-01. Three NVMe slots, four network ports, and a PCIe slot.
• Heavy VM density: Minisforum MS-A2. Sixteen cores and thirty-two threads, with the running cost that implies.

If you want one answer: the MS-01 satisfies both of the constraints below without also charging you for sixteen cores you may not need.

Check the MS-01 on Amazon↗

What Proxmox actually asks of the hardware

Three sentences from the Proxmox VE documentation do most of the work in this comparison.

On networking: “We recommend a dedicated physical NIC for the cluster traffic.” The reason given is not bandwidth. Corosync, the layer that keeps cluster members agreeing with each other, is sensitive to latency jitter, and a link also carrying storage replication or VM traffic is exactly what can introduce it. The documentation adds that “a dedicated 1 Gbit NIC is enough in most situations” — so this is a question of how many ports you have, not how fast they are.

On cluster size: “If you are interested in High Availability, you need to have at least three nodes for reliable quorum.”

On the two-node case: “For smaller 2-node clusters, the QDevice can be used to provide a 3rd vote.” A QDevice is a small external daemon — a Raspberry Pi or an existing always-on box will do — that breaks ties. This matters because buying two mini PCs is the most natural upgrade path, and two nodes on their own cannot hold quorum when one goes down.

None of that is exotic. But it turns two spec-sheet numbers into pass/fail criteria:

• A machine with one network port cannot give corosync a dedicated link. It will still cluster. It will just do so in a configuration the documentation advises against.
• A machine with two NVMe slots cannot hold a ZFS mirror plus a separate boot device. You either boot from the mirrored pair or give up the redundancy you installed ZFS to get.

The NVMe arithmetic is worth spelling out, because it is the constraint people hit after the machine is already on the shelf. A mirrored pair for VM storage uses two slots. Proxmox itself has to boot from somewhere, and while it can boot from a ZFS mirror, that ties the disks holding your guests to the disk holding the hypervisor — replacing one means touching both roles at once. Three slots let you keep those jobs apart: a modest boot drive, and two disks doing nothing but mirroring guest storage. Two slots force a choice. Neither answer is wrong, but the machine makes it for you, and it makes it permanently.

Memory has the same shape. Proxmox will run in 4GB, so the ceiling is really a guest budget: a couple of Linux containers might take 512MB each, while a Windows VM or anything running a database wants 4–8GB to itself. Sixteen gigabytes buys you a handful of containers and one serious VM — a perfectly good home server, and considerably less than what most people picture when they install a hypervisor.

How we compared

Specifications come from the manufacturers’ own product pages. Where a manufacturer does not publish a figure — as with two of the four memory ceilings — it comes off the processor spec sheet instead, and the table footnotes say which is which. The EQ14’s single 16GB slot and the MS-A2’s two slots totalling 96GB are board figures; the other two are not, so read those as upper bounds rather than promises.

Idle power comes from published independent measurements, not from vendor claims, and every figure carries the conditions it was measured under — a number without its test conditions is not comparable to another number.

Where a machine has no published independent measurement, the table says so rather than substituting an estimate.

Three-year electricity is calculated from the idle figure at the U.S. average residential rate published by the EIA. No dollar prices are listed here — the table shows a rough price band, and the buttons go to the current Amazon listing.

We do not own these machines. Everything below is drawn from published specifications and from reviews by people who measured them.

The machines

Beelink EQ14 — the one that is cheap to leave on

The N150 is a four-core, four-thread part in the 6W class, and the EQ14 pairs it with a single DDR4 SO-DIMM slot capped at 16GB. That ceiling is the story. Proxmox itself is happy in 4GB; the VMs are what fill memory, and 16GB fills quickly once you are running more than a few.

What is easy to miss is that it has two 2.5GbE ports. For a machine at this price, that is the difference between a node that can eventually join a cluster properly and one that cannot. If the plan is “one small always-on box now, maybe a second later,” this is the cheapest way to keep that option.

No one has published an independent idle measurement for the EQ14 that we could verify, and Beelink does not publish one either. The N150 is a 6W-class part and the machine very likely sits in single digits, but we are not going to print a number we cannot source.

Check price on Amazon↗

Beelink SER8 — the efficient one with one port

Eight Zen 4 cores, sixteen threads, DDR5, and the lowest idle draw in this group: ServeTheHome measured 7–10W at the wall. For a single-node Proxmox host that runs containers and a few VMs, that combination is hard to beat. Three years of electricity at that draw comes to $34 to $48.

Two things need saying plainly.

It has one 2.5GbE port. For a single node that is fine — one port is all a single node needs. But it means the SER8 cannot follow the documentation’s networking recommendation if you later cluster it, and the upgrade path is a USB adapter, which is exactly the kind of link you do not want corosync depending on.

Its memory ceiling is not what the number suggests. Beelink publishes a maximum of 256GB. The board ServeTheHome reviewed has two SO-DIMM slots, fitted with two 16GB modules, and two SO-DIMMs do not add up to 256GB at any capacity sold today. What you can actually install is the largest pair of SO-DIMMs you can buy — which is well short of what the spec line implies when you read it next to the MS-A2’s 96GB.

Check price on Amazon↗

Minisforum MS-01 — the one built for this

Three M.2 NVMe slots, two 10GbE SFP+ ports, two 2.5GbE ports, and a PCIe 4.0 slot. Read that against the two rules above and it passes both without argument: a boot drive plus a mirrored pair, and four network ports to separate corosync from everything else.

It is sold barebone — chassis, board and CPU — so memory and storage are on you. Minisforum does not publish a maximum memory capacity for it; the 64GB figure is the Max Memory Size Intel lists for the i5-12600H, and the board takes two DDR5-5200 SO-DIMMs. Same caveat as the SER8, in the other direction: that number describes the processor, not the slots.

Power is where it gets interesting, and where the published figures diverge sharply. Scott Stuff measured 13.2W idle on an i5-12600H unit running Ubuntu 24.04 with no add-in network card. ServeTheHome measured 25–29W on their i9-13900H review unit. Both are real. Which one describes your machine depends on the CPU and what you plug into that PCIe slot — a point worth holding on to, because the slot is a large part of why you would buy this machine, and using it is not free.

Check price on Amazon↗

Minisforum MS-A2 — density, and the bill for it

Sixteen Zen 5 cores and thirty-two threads, up to 96GB of DDR5, three NVMe slots, the same four-port networking as the MS-01, and a PCIe 4.0 x8 slot that supports splitting. Against the MS-01’s i5-12600H, that is four more cores and twice the threads.

ServeTheHome measured 25W idle, with one of their configurations idling at 45W. That is the widest spread in this comparison and it is not a rounding difference: at the U.S. average rate, the gap between 25W and 45W is around $32 a year, every year, before the machine has done anything.

If you are consolidating several VMs onto one box, that may be the correct trade. If you are running a handful of containers, you are paying for cores that sit idle while the fan runs.

Check price on Amazon↗

Compared on what Proxmox needs

Specifications from manufacturers. Idle figures from published independent measurements; see the notes for the conditions each was taken under. Power cost assumes continuous operation at the U.S. average residential rate, 18.34¢/kWh (EIA, June 2026).

Model	CPU	Max RAM (GB)	NVMe slots	Networking	PCIe slot	Idle draw (W)	3-yr power cost	Price	Link
Beelink EQ14Beelink	Intel N150 (4C/4T)	16	2	2x 2.5GbE	None	—1	—	$	Check price↗
Beelink SER8 (Ryzen 7 8845HS)Beelink	AMD Ryzen 7 8845HS (8C/16T)	2563	2	1x 2.5GbE	None	7-102	$34–$48	$$	Check price↗
Minisforum MS-01 (i5-12600H, barebone)Minisforum	Intel Core i5-12600H (12C/16T)	645	3	2x 10GbE SFP+ / 2x 2.5GbE	PCIe 4.0 x16 slot (x8 electrical)	13.24	$64	$$$	Check price↗
Minisforum MS-A2 (Ryzen 9 9955HX, barebone)Minisforum	AMD Ryzen 9 9955HX (16C/32T)	96	3	2x 10GbE SFP+ / 2x 2.5GbE	PCIe 4.0 x8 (x16 slot, splitting supported)	25-456	$120–$217	$$$	Check price↗

Read the NVMe and networking columns first. They are the two that rule machines out. Everything else is a matter of degree — more cores, more memory, a bigger bill — but a second NVMe slot does not become a third, and one network port does not become two.

Which one, for what you’re building

Rows are what you are building, not what you are spending.

Use case	If the power bill matters most	If capability matters most
Single node, a few containers and VMs	Beelink EQ14	Beelink SER8 (Ryzen 7 8845HS)
Single node with a ZFS mirror	Minisforum MS-01 (i5-12600H, barebone)	Minisforum MS-01 (i5-12600H, barebone)
Two-node cluster (plus a QDevice)	Beelink EQ14	Minisforum MS-01 (i5-12600H, barebone)
Three-node cluster or Ceph	Minisforum MS-01 (i5-12600H, barebone)	Minisforum MS-A2 (Ryzen 9 9955HX, barebone)
Passthrough: GPU, HBA or a NIC	Minisforum MS-01 (i5-12600H, barebone)	Minisforum MS-A2 (Ryzen 9 9955HX, barebone)

The SER8 appears once, in the top-right cell, and that placement is deliberate. It is the best single-node machine here on performance per watt. It is also the one machine in this comparison that cannot give corosync its own port, so it does not appear in any cluster row.

The EQ14 in the two-node row is the cheap answer, not the comfortable one. Two ports satisfy the networking recommendation, which is the part most budget machines fail, but four threads and 16GB per node put a hard limit on what those nodes can actually host. It is also there on the strength of a 6W-class processor rather than a measurement — its running-cost cell is a dash for the same reason its idle cell is.

The row below it is where the EQ14 stops working, and the reason is Ceph rather than the node count: three nodes ask no more of the network than two, but Ceph wants memory per OSD and storage bandwidth that four threads and 16GB cannot supply.

The MS-01 filling the middle of the table is not padding. Three NVMe slots and four network ports are exactly what the middle rows require, and it is the only machine here that has both without moving up to sixteen cores.

The bottom row deserves a caveat. PCIe passthrough on consumer hardware depends on how the board groups devices under IOMMU, and that varies by firmware version in ways no specification sheet records. None of our sources tested passthrough on these machines. Treat the bottom row as “has a slot to pass through”, not as “passthrough is known to work”.

Frequently asked questions

Can I run Proxmox on a mini PC at all?

Yes. The requirement is a 64-bit CPU with virtualization extensions — Intel VT-x or AMD-V — and enough memory for the guests. All four machines here clear that easily. The constraints that bite are memory, NVMe slots and network ports, not CPU capability.

Do I need three machines to build a cluster?

For reliable quorum with High Availability, the Proxmox documentation says at least three nodes. Two nodes will form a cluster, but they cannot hold quorum when one is unavailable, which defeats most of the reason for clustering. The documented way around it is a QDevice supplying a third vote — that can run on hardware you already have, so a two-node cluster plus a QDevice is a legitimate build, not a workaround.

Why does one network port matter if I only have one machine?

For a single node it does not. It matters when you add the second machine, because the documentation recommends a dedicated physical NIC for cluster traffic and a one-port machine has nothing to dedicate. If you are confident you will never cluster, the SER8’s single port costs you nothing.

How much does one of these cost to run?

The three-year column in the table above does the arithmetic per machine, and our general home server comparison works through the method in more detail.

Does a cluster need matching machines?

No. Proxmox will cluster machines with different CPUs and different amounts of memory. What it will not do is hide the difference. On mixing vendors the documentation is blunt: “Live migrations between Intel and AMD host CPUs have no guarantee to work.” Within one vendor it recommends setting the VM CPU type to the lowest generation in the cluster, which costs you the newer instructions on the faster nodes. Two of the machines here are Intel and two are AMD, so a mixed pair is a real possibility rather than a hypothetical — match the nodes that will share a workload, and let the odd one out be the QDevice.

Should I buy used enterprise hardware instead?

For Proxmox specifically, the case for used enterprise gear is the two constraints this article is built around. A rack server gives you enough PCIe lanes and drive bays that the NVMe arithmetic stops mattering, and enough network ports that corosync gets its own without you having to think about it. If ECC memory matters to you for ZFS, it is also a platform that documents supporting it — neither manufacturer here publishes ECC support for these four machines either way, so if that is a hard requirement, this is not the category to shop in.

The trade is power and noise. A used tower or rack server typically idles somewhere between 40W and 100W, which at U.S. average rates is $64 to $161 a year, every year. That is a general range rather than a measurement we can cite, but it is enough to catch up with the purchase price faster than most people expect.

If you haven’t settled on Proxmox yet

This comparison assumes the hypervisor decision is made. If it is not, the specifications that matter shift — a machine that is wrong for a Proxmox cluster can be right for a single-purpose NAS or media box.