-

DG Matrix Reveals Interport SST for the NVIDIA MGX Ecosystem, Delivering 800 VDC Power for AI Factories

DG Matrix’s multi-port solid-state transformer integrates seamlessly with NVIDIA’s modular MGX architecture, delivering high-density, 800 VDC power for next-generation AI factory deployments

SAN FRANCISCO--(BUSINESS WIRE)--DG Matrix, the global leader in solid-state transformer solutions, today introduced the Interport™ solid-state transformer (SST) platform, designed to integrate seamlessly with the NVIDIA MGX modular rack architecture. The Interport SST delivers native 800 VDC power to GPU-dense racks, positioning DG Matrix at the center of the industry’s accelerating shift to 800 VDC power distribution for next-generation AI deployments.

DG Matrix demonstrated the Interport SST alongside other modular building blocks powering the next generation of AI factories at NVIDIA GTC 2026. NVIDIA MGX is an open rack system architecture comprising standardized, modular compute, networking, and power components designed to accelerate the deployment of NVL72 racks at scale. The MGX architecture features 800 VDC distribution as the path to increasing compute density for performance, reducing conversion losses, and simplifying rack-level power for AI workloads.

The DG Matrix Interport SST is purpose-built for the NVIDIA MGX architecture. The platform natively supports 800 VDC today. Reprogrammable ports support higher voltages of up to 1,500 VDC, enabling future-proofing of today’s data center deployments for upcoming needs. DG Matrix will collaborate with NVIDIA and MGX integrators to validate rack power interfaces and demonstrate MGX-compatible deployments, including 800 VDC rack reference designs.

Since DG Matrix’s founding in 2023, the company has worked closely with NVIDIA on technical alignment around 800 VDC architectures and MGX compatibility, from prototype through the commercial launch of its Interport and Interport Cell products. DG Matrix has also conducted a yearlong program of testing energy storage solutions for dynamic GPU load response in collaboration with NVIDIA.

As AI workloads drive unprecedented electricity demand, power infrastructure has emerged as one of the most critical bottlenecks to deployment speed and efficiency. The DG Matrix Interport platform addresses this challenge directly, delivering up to 98.5% conversion efficiency, rack-proximate power with integrated battery storage, and dynamic GPU pulse-load response in a modular form factor optimized for high-density AI compute environments. The platform’s single-stage, multi-port architecture eliminates complex conversion topologies, reducing both physical footprint and energy losses at scale.

“The Interport SST brings modular, rack-proximate power that delivers the efficiency and dynamic response AI factories require—now validated for the NVIDIA MGX architecture,” said Haroon Inam, CEO and co-founder of DG Matrix. “Collaborating with NVIDIA and MGX integrators helps customers adopt 800-VDC architectures with lower loss and higher power density. Our multi-port capability enables integrated behind-the-meter power aggregation from multiple AC and DC sources along with simultaneous AC and DC outputs, offering a future-proofed platform today. This is a pivotal moment for the industry.”

As part of the NVIDIA MGX ecosystem, DG Matrix will collaborate with NVIDIA and MGX integrators to validate rack power interfaces and advance 800 VDC reference designs for hyperscale, neocloud, and edge deployments for AI training and inference.

About DG Matrix

DG Matrix has commercialized the world’s first multi-port solid-state transformer to solve the most urgent challenges in deploying power infrastructure for AI datacenters and electrification. Its standardized Interport platform enables faster deployment, lower energy costs, and flexible integration of all energy sources and loads—at scale, anywhere in the world. Visit DGMatrix.com for more information.

Contacts

Media Contact:
Will Kruisbrink
ANDOR PR
Will.Kruisbrink@andorpr.com
(217) 766-7894

DG Matrix



Contacts

Media Contact:
Will Kruisbrink
ANDOR PR
Will.Kruisbrink@andorpr.com
(217) 766-7894

Social Media Profiles
More News From DG Matrix

DG Matrix and Satterfield & Pontikes Partner to Accelerate Optimized AI Data Center Construction in Texas

HOUSTON & MORRISVILLE, N.C.--(BUSINESS WIRE)--DG Matrix, the global leader in solid-state transformer solutions, and Satterfield & Pontikes Construction (S&P), the ENR Top 400 general contractor with offices across Texas, today announced a strategic partnership to integrate DG Matrix's Interport™ solid-state transformer (SST) platform into S&P's builds as the general contractor expands its multi-GW data center construction business. The partnership gives S&P priority access to D...

DG Matrix Appoints Christian Belady, Originator of PUE and Former Head of Microsoft’s Data Center Business, to Executive Advisory Board and as Personal Advisor to the CEO

RALEIGH-DURHAM, N.C.--(BUSINESS WIRE)--DG Matrix, the global leader in solid-state transformer solutions, today announced the appointment of Christian Belady, one of the most influential figures in the modern data center industry, to its newly formed Executive Advisory Board. Belady will also serve as a personal advisor to DG Matrix CEO Haroon Inam and work directly with the company’s engineering, solutions and commercial teams as DG Matrix builds the intelligent power fabric for the AI data ce...

DG Matrix Selects Infineon as a Key Silicon Carbide Supplier for Its Multi-Port Solid-State Transformer Platform

MORRISVILLE, N.C.--(BUSINESS WIRE)--DG Matrix, the global leader in solid-state transformer solutions, today announced that it will source latest-generation silicon carbide (SiC) MOSFETs from Infineon Technologies AG (FSE: IFX / OTCQX: IFNNY), a global semiconductor leader in power systems and IoT, for use in its Interport™ multi-port solid-state transformer platform. The relationship strengthens DG Matrix’s semiconductor supply chain and enhances the efficiency, power density, and reliability...
Back to Newsroom