TECH NEWS – In return, Sam Altman’s company will integrate Nvidia’s next-generation Vera Rubin AI platform into its infrastructure.

Demand for AI performance keeps rising, and Nvidia is securing billion-dollar partnerships to expand its computing power and deliver cutting-edge solutions. Just a week after signing a deal with Intel, Jensen Huang’s company announced it will provide $100 billion worth of computing capacity to OpenAI, enabling the deployment of at least 10 gigawatts of Nvidia systems for its next-generation AI infrastructure.

The collaboration centers on Vera Rubin. OpenAI plans to acquire Rubin AI clusters, either the NVL144 configuration or the new Rubin CPX platform. This makes OpenAI one of Rubin’s exclusive customers, granting access to one of the most powerful AI systems available. For over a decade, Nvidia and OpenAI have pushed each other forward—from the first DGX supercomputer to the breakthrough of ChatGPT. This 10-gigawatt investment marks the next leap in AI advancement.

Scale, production, and long-term goals

Energy consumption estimates suggest this deal could result in up to 40,000 Rubin AI racks sold to OpenAI—a staggering figure for a product not yet in mass production. The Rubin lineup is scheduled to enter mass production in the second half of 2026, with OpenAI and other companies gaining access simultaneously. Rubin represents a major step forward compared to Blackwell.

The contrast with Intel is striking. For Intel, Nvidia’s investment was essentially a lifeline, while for OpenAI, it’s a windfall. Long-term, Nvidia aims to integrate 5G modem, Wi-Fi, and Bluetooth chips into one package to save space and extend efficiency.

Source: WCCFTech, Nvidia