Amazon and Nvidia are getting much closer. The two companies announced an expanded partnership Wednesday that includes an agreement to add another 2 million Nvidia GPU chips to Amazon data centers. These GPUs, which are designed to handle the large computing demands of training and running AI models, include the Nvidia Blackwell Ultra, Rubin, and Rubin Ultra GPUs. The chips will head to Amazon Web Services data centers in 2027 and 2028. The announcement, made during Nvidia’s quarterly earnings call, comes just five months after Amazon agreed to deploy more than 1 million Nvidia GPUs on AWS infrastructure starting this year. Nvidia said in a statement that since then, “demand has exceeded those expectations.” Neither company shared financial terms. It’s unclear what the exact return will be for Nvidia. But considering GPU unit costs, the deal is worth tens of billions of dollars. The announcement is notable not only for its size and the speed with which it grew, but also because it goes beyond Amazon buying more Nvidia chips. And it’s happening even as Amazon invests in its own potentially competitive AI chips. Nvidia said Wednesday that its technology, including networking hardware that connects thousands of GPUs into a single system, as well as its open models, CPUs, data processing software and robotics platform, will also be integrated into AWS. The companies said “growing demand” from startups, companies, AI labs and even governments influenced the decision to work more closely. The expanded partnership comes as Amazon steps up its own efforts in AI chips, particularly with CPUs, which are the general-purpose processors at the heart of servers. Amazon has been building its own chips to lessen its dependence on Nvidia and even compete with the chip giant. Amazon AI chief Peter DeSantis has said AWS is in talks to sell its Trainium chips, which are a direct alternative to Nvidia’s H100 or Blackwell chips for deep learning workloads, to other companies for use in data centers. The Graviton CPU built by Amazon’s Arm is also seen as a challenger to traditional server chips from Intel and AMD. Amazon has said its custom chip business is growing, noting in its latest earnings call that it surpassed a $25 billion annualized revenue rate, driven by $225 billion in total commitments from AI labs such as Anthropic and OpenAI. But it seems that Nvidia is still the GOAT in the world of AI chips. With the 2 million GPU chips Amazon is adding to AWS starting in the third quarter, Nvidia also plans to ship an unspecified number of Vera CPUs, “some integrated with Rubin, some standalone,” according to Nvidia CFO Colette Kress. Nvidia CEO Jensen Huang has big plans for the company’s Vera CPUs, and in May he boasted that he had found a “new $200 billion TAM” for the company. In addition to AWS, Kress said Wednesday that Nvidia expects Vera to be deployed by “all major hyperscalers, neocloud, AI labs and system OEMs, with shipments already underway to our core partners,” which include Oracle and SpaceXAI. The partnership is also extending to Amazon’s warehouse robots and enterprise offerings. Kress said Amazon plans to adopt Nvidia’s full physical AI stack to power its robot fleet. The stack includes Omniverse (its digital twin and simulation platform); Cosmos (its world model platform); Isaac (its robotics development platform); and Jetson (computer hardware for cutting-edge robots and AI). This week, Nvidia also introduced a new version of Jetson, designed as a more accessible robotic computer for “entry-level AI.” On the enterprise side, AWS will serve Nvidia’s Nemotron family of open models on Amazon Bedrock, its managed core model platform, and SageMaker, its managed cloud service. Nvidia also reported Wednesday that it posted sales of $96.2 billion in the second quarter, beating analyst estimates. Data center revenue accounted for the majority of Nvidia’s sales during the quarter at $89 billion, up 117% from a year ago. Nvidia said it expects revenue to reach $108 billion in the third quarter, some of which will come from its next-generation Rubin GPUs. Nvidia said it began production shipments this quarter. Investors have been watching Rubin’s initial third-quarter sales for signs that demand will continue for Nvidia’s next generation of hardware. Nvidia has committed $279 billion to secure supply and manufacturing capacity for current and future data center projects, substantially up from $119 billion last quarter, as the chipmaker looks to secure memory and manufacturing capacity to meet AI demand for years to come. That commitment includes $92 billion in projected spending for the rest of the fiscal year and another $87 billion in fiscal 2028. “What matters for the industry is that AI is now doing productive and useful work,” Huang said during Wednesday’s call. “AI is generating profitable tokens… If we had more computing, we could generate more profitable tokens, resulting in more profits for all services. This is the exact phase we are in, which is why everyone is leaning towards it.” Investors will be watching to see whether additional computing translates so clearly into additional profits as AI companies invest hundreds of billions of dollars in infrastructure. When you buy through links in our articles, we may earn a small commission. This does not affect our editorial independence.
