NVIDIA official newsroom (nvidianews.nvidia.com/bios/jensen-huang)
Nvidia qualifies Tesla and Vertiv for new AI-factory standard
The chipmaker's DSX Ready program creates a certified supply chain for the massive power and cooling demands of AI data centers.
Nvidia is no longer content with just selling the brains of AI data centers. Now it wants to certify the plumbing and the power grid, too.
The company launched its DSX Ready qualification program, a certification framework designed to vet third-party power and cooling products for use in its DSX AI factory reference designs. Tesla, Hitachi Energy, and LG Energy Solution earned qualification for their battery energy storage systems, while Vertiv, LG Electronics, and LiquidStack were certified for their cooling distribution units.
What DSX Ready actually does
The DSX Ready program evaluates products across two categories: battery energy storage systems (BESS) and cooling distribution units (CDUs). For the BESS category, partners submit test data that Nvidia evaluates against its reference design specifications. CDU qualification follows a self-certification model. In both cases, the certification signals compatibility with Nvidia’s blueprint for AI infrastructure, though the company is careful to note that qualification doesn’t guarantee performance at any specific deployment site.
Nvidia has signaled plans to expand the program into additional infrastructure and software categories in future phases.
Why Tesla and Vertiv matter here
Tesla’s inclusion comes through its Megapack product, the company’s grid-scale battery storage system. These facilities consume enormous amounts of electricity, and they need backup storage systems that can handle sudden demand spikes.
Macro, rates, and crypto—what moved markets and what matters next.
Daily. Free. Join 34,000+ readers across crypto, finance, and policy.
Vertiv’s certification builds on what was already a deep relationship with Nvidia. The company has been collaborating with Nvidia on prefabricated systems and digital-twin modeling for AI factory designs, meaning this qualification formalizes a partnership that was already producing real products.
The bigger picture for AI infrastructure
A single rack of Nvidia’s latest GPUs can draw more than 100 kilowatts, and modern AI clusters stack hundreds or thousands of these racks together, with entire facilities scaling to hundreds of megawatts. The cooling requirements have pushed the industry toward liquid cooling solutions, as air cooling cannot dissipate heat fast enough for the density of modern AI hardware.
By standardizing the evaluation of power and cooling components, Nvidia is positioning itself as the architect of the entire AI data center stack, not just the compute layer. Certification doesn’t guarantee revenue, but it does reduce friction in sales conversations, and choosing a DSX Ready component represents the path of least resistance when assembling an AI factory to Nvidia’s specifications.