Nvidia on Monday unveiled Vera Rubin, a transformative computing platform built from seven advanced chips and now in full production. This platform is backed by an impressive roster of clients including Anthropic, OpenAI, Meta, and Mistral AI, alongside every major cloud provider.
The message to the AI industry is clear: Nvidia is not slowing down. The Vera Rubin platform promises up to 10x more inference throughput per watt and one-tenth the cost per token compared to its predecessor, the Blackwell systems. CEO Jensen Huang, during the company’s annual GTC conference, termed it "a generational leap" that will initiate "the greatest infrastructure buildout in history." Major cloud services like Amazon Web Services, Google Cloud, Microsoft Azure, and Oracle Cloud Infrastructure plan to offer this platform, with over 80 manufacturing partners developing systems around it.
“Vera Rubin is a generational leap — seven breakthrough chips, five racks, one giant supercomputer — built to power every phase of AI,” Huang asserted. “The agentic AI inflection point has arrived with Vera Rubin kicking off the greatest infrastructure buildout in history.”