Nvidia expects to sell $20 billion of Vera Rubin systems in Q3, representing 20% of data center revenue and marking its fastest historical ramp.

Nvidia expects sales of its Vera Rubin platform to account for approximately 20% of its data center revenue in the third quarter of fiscal year 2027, marking the fastest product ramp in the company’s history. With total projected revenue for Q3 FY2027 standing at around $108 billion, Vera Rubin hardware alone is expected to generate roughly $20 billion in a single quarter. Given that data center sales have consistently represented about 92% of Nvidia’s quarterly revenue in recent periods, meeting the $108 billion projection would yield approximately $99.36 billion in data center revenue, translating to roughly $19.872 billion specifically from Vera Rubin platform hardware.

“We see Vera Rubin accounting for about 20% of data center revenue in Q3,” said Colette Kress, chief financial officer at Nvidia, during the company’s earnings conference call with financial analysts and investors. “Having already received purchase orders from every major hyperscaler, AI cloud, and system OEM, we expect Vera Rubin to mark the fastest product ramp in Nvidia’s history.”

Nvidia began ramping production of Vera Rubin components—including Vera CPUs, Rubin GPUs, BlueField 4 DPUs, and other units—this spring. First revenue shipments of complete VR200 NVL72 racks commenced in August, with Microsoft becoming the first customer to deploy commercial systems, according to CEO Satya Nadella. Typically, companies require multiple quarters to scale production and sales for new data center platforms. Accelerating from effectively zero production revenue in Q2 FY2027 to capturing 20% of quarterly revenue ($20 billion) in a single quarter represents a record not only for Nvidia but potentially for the entire industry, given the company’s market scale.

There is, however, a caveat: actual unit shipment volumes do not reflect a massive breakthrough in quantity. While exact shipment figures for Q3 FY2027 remain undisclosed, estimates based on pricing suggest otherwise. Assuming Nvidia ships exclusively VR200 NVL72 rack-scale systems, and considering an estimated price range of $5 million to $7.8 million per unit, the company would supply between approximately 2,550 and 3,975 machines. Each configuration would contain roughly 91,700 to 143,100 Vera CPUs and 183,500 to 286,100 Rubin GPUs.

These figures are necessarily rough estimates, as Nvidia will also distribute MGX servers and sell Vera CPUs and Rubin GPUs individually, which would increase overall component shipment volumes. Nevertheless, the totals still amount to hundreds of thousands rather than millions of processors and accelerators—a relatively modest volume for leading-edge AI hardware. However, due to the premium pricing of cutting-edge AI infrastructure, even these constrained shipment numbers are sufficient to generate tens of billions of dollars in revenue.