Nvidia offered the latest view of what’s to come for the chipmaking giant today, with CEO Jensen Huang showcasing Blackwell Ultra and subsequent generations of AI chips at a company conference.
Nvidia’s Blackwell Ultra GB300 family of semiconductors will have one-and-a-half times the memory and deliver significantly higher performance than the current Blackwell line, Huang said at the company’s GPU Technology Conference.
Nvidia expects Blackwell Ultra to launch in the second half of 2025.
Today’s news sets the stage for Nvidia’s next-generation CPU-GPU platform, known as Vera Rubin, scheduled for the second half of 2026, and Rubin Ultra GPUs a year later, Huang said.
Blackwell has 68 times the computing power of its Hopper line, he said, while Rubin represents a 900-fold increase. (Vera Rubin is the astronomer who discovered dark matter; a later line, Huang said, will be named for physicist Richard Feynman.)
The announcements included a partnership with General Motors focused on developing driverless vehicles that would feature an Nvidia-made, in-vehicle computing system that can deliver up to 1,000 trillion operations per second.
He also unveiled a telecoms project, involving T-Mobile and Cisco Systems, where Nvidia will help create AI-ready hardware for wireless 6G networks, the successor to today’s 5G.
Huang on Tuesday pointed to surging demand for Nvidia GPUs. The company shipped 3.6 million Blackwell GPUs to America’s four largest cloud service providers—Microsoft, Alphabet, Amazon, and Meta—in its 2025 fiscal year, which ended in late January. That’s up from 1.3 million Hopper GPUs the year before.
Nvidia reported it finished last year with record high revenue of $130.5 billion, driven by demand for its chips to power AI in data centers.
Nvidia projected revenue of $43 billion in the current fiscal quarter, topping analyst expectations.
Follow the MoneyCentral Nigeria channel on WhatsApp: Click here!
