Skip to content
EMARQUE.AI
    HomeProductsNVIDIA Vera Rubin NVL72
Roadmap — allocation openNVIDIA DGX

NVIDIA Vera Rubin NVL72

The next-generation NVIDIA rack-scale AI platform — NVIDIA Vera CPUs + NVIDIA Rubin GPUs with HBM4 memory and NVIDIA ConnectX-9 networking. Announced for H2 2026.

Request allocationCompare systems
Pricing on request — allocation and configuration confirmed at quotation.
NVIDIA Vera Rubin NVL72 — built by EMARQUE in Malaysia
RubinGPU generation
HBM4Memory
~3.6ExaFLOPS FP4 inference
Two ways to buy

Same platform — choose the supply path.

The NVIDIA HGX baseboard is identical on both paths. The turnkey DGX is fastest to deploy; OEM HGX platforms (Dell, Giga Computing, Supermicro) give wider configuration choice.

NVIDIA

NVIDIA DGX Vera Rubin (turnkey)

NVIDIA-built and NVIDIA-supported packaging of the NVIDIA Vera Rubin NVL rack platform — full NVIDIA DGX OS, NVIDIA Mission Control, and NVIDIA Enterprise Support contract included. Continues the NVIDIA DGX SuperPOD operating model into the Rubin generation.

  • NVIDIA DGX OS · NVIDIA Mission Control · NVIDIA AI Enterprise (latest)
  • NVIDIA Enterprise Support contract included
  • Reference architecture for next-generation NVIDIA DGX SuperPOD
  • Operational continuity from existing NVIDIA DGX GB300 estates
Giga Computing · Supermicro · Dell · ASUS

NVIDIA Vera Rubin NVL — OEM platforms

The same NVIDIA Vera Rubin NVL rack platform from NVIDIA's OEM partners — customer-operated, no DGX software bundle. Roadmap availability; OEM partner model numbers confirmed closer to GA.

  • Customer-operated orchestration (Slurm, Kubernetes, OpenStack)
  • OEM warranty (Giga Computing, Supermicro, Dell — confirmed at quotation)
  • Lower total cost without the DGX software-bundle premium
  • Allocation slot reserved for late-2026 / 2027 deployment window

EMARQUE supplies both paths in Malaysia. Final configuration, lead time, and warranty terms are confirmed in writing at quotation.

Next step

Get a Vera Rubin NVL72 configuration and lead time from your Malaysian NVIDIA systems specialist.

Talk to a Vera Rubin NVL72 specialistCompare systemsContact us
Full spec sheet

Every line documented at quotation.

As supplied by NVIDIA. EMARQUE handles in-country delivery, commissioning, and Tier-1 support handoff.

Status
Roadmap — NVIDIA announced at GTC 2025. Allocation conversations open.
GPU
NVIDIA Rubin GPU (dual-die)
CPU
NVIDIA Vera CPU
Memory
HBM4 (capacity per GPU per NVIDIA's GA datasheet)
FP4 compute
~3.6 ExaFLOPS inference per rack (NVIDIA GTC 2025)
FP8 compute
~1.2 ExaFLOPS training per rack (NVIDIA GTC 2025)
NVLink fabric
Next-generation NVLink Switch (NVIDIA announced)
Networking
NVIDIA ConnectX-9 SuperNICs · 1.6 Tb/s per GPU
Cooling
Direct-to-chip liquid
Target availability
H2 2026 — confirm at allocation
Also in this class
NVIDIA GB300 NVL72
Rack-scale AI infrastructure for the era of reasoning — 36 NVIDIA Grace CPUs and 72 NVIDIA Blackwell Ultra GPUs in a single NVLink-connected, liquid-cooled rack.

Request configuration & quotation.

Manufacturer specifications, factory lead times, and warranty terms apply. EMARQUE responds within one business day with a formal quotation and projected delivery window.

Request allocationContact sales
Contact Us

Get in Touch with Us

Tell us about your workload. We reply within one business day with a quote sized to fit.

Build your workstationContact us
  1. 01

    Key Account Manager

    +6012 627 2280
  2. 02

    Request for Quotation

    business@emarque.co