ASRock Rack 2U1G-B650/EVAC — 2U Single-GPU Barebone, Short Depth, 1+1 1200 W
Enterprise-grade platform, assembled and tested in the EU.
ASRock Rack 2U1G-B650/EVAC
One full-size GPU, redundant power, 478 mm deep
The entry rack GPU node — and the one that fits a shallow cabinet.
A 2U node on AMD’s AM5 socket that takes one full-length, dual-slot GPU and backs it with a genuinely redundant 1+1 1200 W Platinum supply. At 478 mm deep it goes into shallow and wall-mount cabinets that will not accept a normal 850 mm server — which is often the actual constraint in a comms room or at an edge site.
Specification
| Component | Detail |
|---|---|
| Form factor | 2U rackmount, 477.7 × 430.4 × 88 mm |
| Processor | Single Socket AM5 (LGA 1718). AMD EPYC 4005 / 4004, or Ryzen 9000 / 8000 / 7000. Up to 170 W TDP. Chipset B650E |
| Memory | 4 DIMM slots, 2DPC. DDR5 ECC or non-ECC UDIMM, 48 GB per module — 192 GB maximum. 5600 MHz at 1DPC, 3600 MHz at 2DPC |
| GPU slot | 1 × FHFL dual-slot PCIe 5.0 x16 on riser, plus 1 × FHHL PCIe 4.0 x4 |
| Storage | 2 × M.2, 1 × optional fixed 2.5" bay |
| Networking | 2 × 1GbE Intel i210, dedicated IPMI LAN |
| Power | 1+1 hot-swap CRPS, 1200 W at 200–240 V (1000 W at 110 V), 80 PLUS Platinum |
| Management | ASPEED BMC, IPMI |
What it will and will not carry
1200 W redundant against a 170 W CPU leaves real room for the card: a 600 W class GPU fits the budget with margin, and a 300 W Max-Q card leaves the machine barely working. One card is the limit — there is one x16 slot and no second position.
The constraint to watch is memory. AM5 takes UDIMM, capped at 48 GB per module and 192 GB in the chassis. If you size system memory at twice total VRAM, this comfortably carries one 96 GB card and nothing larger. Storage is thin too — two M.2 and one optional 2.5-inch bay — so plan on network storage or keep the dataset on the M.2 pair.
Within those limits it is the cheapest way to put a serious GPU in a rack with redundant power: a single-model inference endpoint, a render or transcode node, a CI runner with a real card in it, or a branch box that has to live in a shallow cabinet.
Rack and cooling
Front-to-back airflow, 24/7 rated. Check the cabinet depth before anything else — the short chassis is the reason to choose this machine, and it is also why it will look lost in a standard 1000 mm rack.
What is included
What is in the box: chassis, backplane, motherboard, the full fan wall and the power supplies listed above, pre-fitted and cabled. It is a complete, working platform less the parts you choose.
What is not: CPU, memory, drives, GPU and any add-in NIC. Those are specified with you and priced separately — that is the point of buying a barebone rather than a finished server.
What Kentino adds
Every barebone we ship is unpacked, assembled, powered up and tested before it leaves Prague: fan wall and PSU redundancy checked, BMC firmware brought up to the current release, IPMI configured and the network details handed over in writing. You get a machine that has already been switched on once by somebody who knows what it should sound like.
Tell us the workload instead of the part numbers and we will specify the CPU, memory and storage for you, build it, burn it in and ship it as a finished server. The same chassis, the same price for the chassis — we simply add the parts and the work.
Price shown is EUR 2,967 excluding VAT for the barebone as described, EXW Prague (Incoterms 2020). Freight, insurance and any destination duties are for the buyer’s account. Typical lead time 10–28 days. Image: ASRock Rack.
The questions buyers ask us most often before ordering a server.
How long does it take?
Machines built from components we hold ship quickly; anything requiring a specific GPU generation depends on supply. We give you a date before you pay, and if it moves we tell you rather than letting you find out.
Can the configuration be changed before you build it?
Almost always. GPUs, memory, storage and cooling are chosen per order, and the listed configuration is a starting point rather than a fixed package. If you need more VRAM, faster storage or a different cooling approach, say so before you order and we will quote the change.
Can I collect the server in person?
You can. Our warehouse is in Prague, and collection in person is welcome — most people who come use the visit to go through the machine with our engineer and ask the questions that are awkward over email. For orders within the Czech Republic we also try to deliver personally and walk you through the setup on site.
Can I talk to someone who actually understands the workload?
Yes. We have an engineer who works on AI systems specifically, not a general sales desk. If your question is about batch sizes, quantisation, interconnect or where your bottleneck will be, ask it — that conversation usually changes the configuration for the better.
Which model can I run on this configuration?
Yes. Every machine is assembled, burn-in tested and benchmarked on real AI workloads before it ships, and it leaves us with an LLM already installed and running. You plug it in, connect it to your network and start work — the only decision left is which project it runs first.
How do you test a server before shipping?
We run it against actual AI workloads rather than synthetic scores: inference throughput, sustained load behaviour and thermals under continuous operation. You get the benchmark results with the machine, so the performance you were promised is the performance you can verify on day one.
Is the server ready to run when it arrives?
Yes. Every machine is assembled, burn-in tested and benchmarked on real AI workloads before it ships, and it leaves us with an LLM already installed and running. You plug it in, connect it to your network and start work — the only decision left is which project it runs first.
Ships from our EU warehouse. Heavy items may require freight arrangement — contact us for a shipping quote and lead time. 2-year limited warranty with advanced RMA support; extended warranty available.
Not exactly what you need?
Tell us your workload and we'll spec this platform around it — GPUs, memory, storage and cooling matched to what you actually run.