Supermicro AS-A126GS-TNBR NVIDIA HGX B200 8-GPU AI Server – 10U, Dual AMD EPYC 9005

The Supermicro AS-A126GS-TNBR combines the NVIDIA HGX B200 8-GPU board with two AMD EPYC 9005 processors in a 10U air-cooled chassis. AMD's 5th-generation EPYC gives you the widest choice of host CPUs: high-frequency parts such as the EPYC 9575F to keep eight Blackwell GPUs fed, or up to 192 cores per socket when the server also runs data processing, virtualisation or inference micro-services.

It is a strong fit for teams standardised on AMD, for LLM fine-tuning and training, multi-tenant inference and HPC workloads that still need FP64. Mineshop.eu configures it to your workload and quotes it with EU delivery, VAT handling and an optional onsite service plan.

Explore AI & GPU Servers and compare AS-8126GS-NB3RT (HGX B300, AMD) with SYS-A22GA-NBRT (HGX B200, Intel).

Key features

• 8× NVIDIA B200 SXM on the HGX B200 board – 1.4 TB HBM3e, NVLink 1.8 TB/s per GPU
• Dual AMD EPYC 9005 (or 9004) – up to 384 cores in total, up to 500 W TDP
• 24 DIMM slots – up to 6 TB DDR5-6400
• 8× hot-swap 2.5" NVMe + 2× SATA bays, optional 8× E1.S (with BlueField-3), 2× M.2 NVMe
• 8× PCIe 5.0 x16 LP + 2× FHHL slots for a 400G AI fabric and DPUs
• 6× 5250 W Titanium PSUs (3+3 redundant), air-cooled 10U chassis

Workloads and use cases

• Fine-tuning and training: Eight B200 GPUs with 1.4 TB of HBM3e and NVLink Switch for full-node data, tensor and pipeline parallelism.
• Multi-tenant inference: High core-count EPYC hosts run Kubernetes, vLLM / TensorRT-LLM serving and RAG services next to the GPUs.
• Scientific computing: 296 TFLOPS of FP64 across the HGX B200 board for simulation and HPC codes.
• AMD-standardised fleets: Same EPYC 9005 platform, tooling and BIOS management as your existing AMD servers.

NVIDIA HGX B200 8-GPU performance

System-level figures published by NVIDIA for the eight-GPU board; real application performance depends on model, precision and software.

AS-A126GS-TNBR full specifications

Configuration options

Every HGX server is configured to order. Typical options are listed below; final compatibility and part numbers are confirmed in your quote.

Delivery and lead time

• Estimated lead time: Confirmed in your quote; custom HGX builds can take up to ~30 weeks depending on GPU allocation.
• Quote: written B2B quote excluding VAT within 1 business day, including the confirmed delivery window.
• Shipping: palletised freight to your data centre in all 27 EU countries; delivery to other countries on request and subject to export-control checks.
• Before shipping: optional burn-in test with report, firmware update and OS pre-installation.

Data-centre checklist

• Power: 6× 5250 W Titanium PSUs in 3+3 redundancy on 200–240 VAC – plan two independent feeds/PDUs. The exact power budget is part of the quote.
• Space: 843 mm deep 10U chassis – a 1000–1200 mm deep rack.
• Weight: 133 kg net – use a lift for installation and check rack and floor load ratings.
• Cooling: front-to-back airflow, inlet air 10–35 °C; hot/cold-aisle containment recommended.
• Network: plan leaf switches, optics and cables for the GPU fabric (8× 400G per server) plus management and storage networks.

Buying from Mineshop.eu

• New systems sourced through European IT distribution, with manufacturer warranty and onsite service options.
• B2B invoicing excluding VAT, EU reverse charge for VAT-registered businesses, prices in EUR.
• Hardware specialists who help you size GPUs, memory, storage and networking for your models.
• One supplier for the complete stack: servers, GPUs, 400G/800G networking, server memory and NVMe storage.

Ordering process

1. Click Request a quote and tell us the configuration, quantity, delivery country and timeline.
2. A hardware specialist confirms the configuration with you – by email or a short call.
3. You receive a written quote within 1 business day, with lead time, service plan and payment terms.
4. You confirm with a purchase order; we handle export-control and end-user paperwork.
5. Your server is built, tested and delivered; optional rack integration and onsite installation.

Compare Supermicro HGX servers

Frequently asked questions

What is the price of the Supermicro AS-A126GS-TNBR?

The AS-A126GS-TNBR is quoted per configuration because B200 GPU prices and allocations move quickly and the CPU, memory and network choices change the total. You receive a written B2B quote, excluding VAT, within 1 business day of your request.

Which AMD EPYC CPU is best for an HGX B200 server?

For GPU-bound training and inference most buyers pick a high-frequency part such as the EPYC 9575F (64 cores, 3.3 GHz base) to keep the GPUs fed. If the server also runs CPU-heavy services, higher core-count EPYC 9005 models up to 192 cores per socket are available.

What is the difference between the AS-A126GS-TNBR and the SYS-A22GA-NBRT?

Both are 10U HGX B200 8-GPU systems. The AS-A126GS-TNBR uses AMD EPYC 9005/9004 CPUs with DDR5-6400 memory; the SYS-A22GA-NBRT uses Intel Xeon 6900P CPUs with DDR5-6400 RDIMM or DDR5-8800 MRDIMM and offers ten U.2 NVMe bays and twelve PCIe slots.

Can I use E1.S drives instead of U.2?

Yes. Next to the 8× 2.5" NVMe and 2× SATA bays, the chassis offers 8 optional hot-swap E1.S NVMe bays. They require an NVIDIA BlueField-3 adapter and additional cabling, which we include in the quote.

What does the server need in my data centre?

A 200–240 V supply for six 5250 W Titanium PSUs (3+3 redundant), 10U in a deep rack, a rack and floor rated for 133 kg and 10–35 °C inlet air. The exact power budget and cords are confirmed with the quote.

How long is delivery?

Lead time is confirmed in your quote; custom HGX builds can take up to ~30 weeks depending on GPU allocation. If you need GPUs sooner, ask about our Gold Series HGX B300 systems (6–12 weeks).

NVIDIA, HGX, NVLink, ConnectX and BlueField are trademarks of NVIDIA Corporation. Supermicro is a trademark of Super Micro Computer, Inc. Specifications from the manufacturers; subject to change without notice.