Buy a server with 4× NVIDIA RTX PRO 6000 Blackwell Server Edition in Europe

Who the four-card server is for

You can buy a server with four NVIDIA RTX PRO 6000 Blackwell Server Edition cards from us, built to order and delivered anywhere in the EU. It is the inference node among our AI servers: four 96 GB cards, two AMD EPYC processors, 768 GB of DDR5 ECC memory and four NVMe drives, for production inference for several teams and several models at once. Our team sizes the configuration to your workload and budget and replies within one business day.

The four-card server suits a company that has finished a pilot and now runs a private assistant, RAG on its documents or coding help for several departments. It holds one large model with room for many conversations, or a 70B model next to an embedding model, a reranker and a speech-to-text service. With MIG it also gives up to 16 isolated GPUs of 24 GB for development teams. For a pilot or one department, two cards are enough; models with more than about 330 GiB of weights need eight cards. Training and full fine-tuning of large models belong on our training node with eight H200 NVL and NVLink bridges.

What the configuration contains

The table shows the reference configuration and what we change per project.

Our AI servers page; NVIDIA-Certified Systems configuration guide (30 September 2026) for cores, memory, Redfish and TPM; HPE DL380a Gen12 QuickSpecs and Cisco UCS C845A M8 data sheet for power-supply ratings, read on 24 October 2026.

The host follows NVIDIA's configuration guide for NVIDIA-Certified Systems: at least two sockets, six physical cores per GPU and system memory of at least twice the total GPU memory, so 768 GB for four 96 GB cards. With two 32-core EPYC 9355, 24 modules of 32 GB fill all 24 memory channels. Four NVMe drives cover a mirrored system and a mirrored model store; the reasoning behind each tier is in our guide to CPU, RAM and storage for a 4-GPU server.

Variants: 2U at 450 W or 4U at 600 W

The power mode decides most of the rest. A 2U server with the cards capped at 450 W takes the least rack space and, by our estimate, stays within one 230 V 16 A feed per side; Lenovo, for example, allows four cards in its 2U ThinkSystem SR650a V4 only in the version "Slot Capped at 450W". StorageReview measured the cost of the cap at 2 to 9 per cent of throughput, depending on context length, on a Workstation Edition card standing in for the Server Edition. A 3U or 4U server runs the cards at 600 W with more room for air, power supplies and network cards, and needs a rack position with a 32 A or three-phase feed.

• Memory. 768 GB covers serving; about 1 TB suits a large vector index or a KV cache offloaded to system memory for many returning users.
• Network. 25 GbE is enough for chat traffic; 100 GbE or more pays off when models load from shared storage or the server joins a cluster.
• Growth. A server can start with two cards in a chassis, power supplies and slot layout chosen for four, as our guide to adding GPUs later explains.
• Max-Q instead of Server Edition. The RTX PRO 6000 Max-Q is the workstation card, a 300 W blower card that NVIDIA supports up to four per workstation. In a rack server it belongs only in a chassis whose maker lists it, and many server makers do not; for continuous operation in a rack we recommend the Server Edition, as our page on the Max-Q in a rack server explains.

What it runs

By our users-per-card rule, 0.9 × 95.6 GiB − 3 GiB per card, the four cards hold 332.2 GiB of weights and KV cache. The table gives four examples; our guide to what a 4× RTX PRO 6000 Server Edition server runs has the full model list, every layout, published results and the arithmetic.

Our estimates by the users-per-card rule: concurrent conversations at full context length, FP8 KV cache for Llama 3.3 70B, 16-bit for the others.

These are requests in flight at the same moment, which a company's gateway logs show at the busiest hour. We supply the server in every variant above; send us your models and peak requests, and our team returns a configuration with the conversations per layout.

Power and room requirements

By our estimate, four cards at 600 W with two 280 W processors need about 3.4 kW at the wall, and about 2.7 kW at the 450 W cap. We plan a 230 V 16 A single-phase feed at 80 per cent, which is 2.94 kW, so the 600 W version needs a 32 A single-phase (5.89 kW planned) or a 400 V three-phase position. With redundant power supplies on an A and a B feed, each side carries the whole server.

• Outlets. HPE, for example, allows only C19 or C19-C20 cords for the 2,400 and 3,200 W supplies its DL380a Gen12 needs with this card, so check that the rack PDU has C19 outlets.
• Cooling. All the power becomes heat: about 3.4 kW needs roughly 540 CFM of air at an 11 °C rise. NVIDIA's product brief asks for 43 CFM through each card's heat sink at a 25 °C inlet and 125 CFM at 45 °C, and server makers set their own inlet limits per configuration.
• Room. A four-card server at full load belongs in a server room or data centre, as our guide to a GPU server in an office explains.

We check the rack, feed and airflow before we quote. Tell us the feed type, the A/B requirement and the inlet temperature at the rack position, and we choose the power mode and chassis to match.

Software delivered with the server

On request the server arrives with the operating system, NVIDIA driver, CUDA and a container runtime installed, so your team can start a serving engine on the first day.

Our AI servers page; driver and CUDA minimums from NVIDIA's product brief SP-12355-001_v02 for the Server Edition.

As of October 2026, R580 is NVIDIA's long-term branch with support to June 2028, and NVIDIA's security bulletin of 30 September 2026 names 580.178.04 as the first fixed R580 driver; our article on NVIDIA driver branches and CUDA versions explains the choice. Bare-metal serving with open-source engines needs no NVIDIA licence. NVIDIA AI Enterprise, licensed per GPU, is needed for NIM and other NVIDIA AI Enterprise software and for compute vGPU profiles; on vSphere the card needs ESXi 8.0 Update 3g or later. Models, RAG and the platform on top are a separate service, Private AI/ML.

How ordering works with us

The order runs in the same steps for one server or several.

1. You describe the workload: models, users, data, rack and timeline. Sizing is free of charge.
2. We propose the configuration and check that cards, chassis, power supplies and cooling fit your rack and feed.
3. You receive a written quote within one business day, with price and warranty terms per component; quantities and lead time are confirmed in writing with it.
4. You order on one EU contract with an EU invoice from Eurokommerz in Vienna; NVIDIA licences are lines on the same invoice.
5. We assemble the server and test it before delivery, with a power-on and basic load test.
6. We deliver anywhere in the EU, with commissioning on site on request; other destinations on request through our logistics partner.
7. Every component carries the manufacturer warranty, handled through us; DOA units are replaced, and the assembly is covered by us.

We also build in your own chassis or with parts you already own, after a compatibility check of the platform, power and cooling. For your own acceptance test on delivery, use the checks in our GPU server acceptance testing guide; to compare our quote with others line by line, our guide to GPU server quote comparison lists what each line should state.

What to send us for a quote

A short description is enough to start; these points let us quote the exact configuration in the first reply.

• Models you plan to run, with quantisation and the context length you will declare
• Peak requests in flight, or the number of users and how they work
• Other workloads on the same server: RAG with embedding and reranking, speech-to-text, fine-tuning, virtual machines or MIG
• Software platform: bare metal, Kubernetes or vSphere, with the version
• Rack position: free height units, feed type and current, A/B feeds, PDU outlets and inlet temperature
• Network uplinks and where models and data are stored
• Delivery site, invoicing entity, timeline, and any tender requirements, which our GPU server tender specification helps to word

Describe these points in the form below, and our team replies with a configuration and a written quote within one business day.

What we supply

We supply the server with four NVIDIA RTX PRO 6000 Blackwell Server Edition cards in 2U, 3U or 4U, at 450 or 600 W per card, together with two-card and eight-card versions and the RTX PRO 6000 cards for servers you already run. NVIDIA AI Enterprise and vGPU licences come on the same EU contract and invoice as the hardware.

FAQ