At the annual CES technology event in Las Vegas last week, Nvidia CEO Jensen Huang announced that the company's next-generation Vera Rubin AI chips are now in full production. This development occurred a full six months ahead of the previously scheduled timeline for the second half of 2026.
"Vera Rubin is designed to address this fundamental challenge that we have: The amount of computation necessary for AI is skyrocketing," Huang stated during his keynote address. "Today, I can tell you that Vera Rubin is in full production."
The announcement surprised industry observers, as Nvidia had initially projected full production for later in 2026. The company reports that the Rubin architecture reduces AI token processing costs by up to 90% while using 75% fewer GPUs.
Nvidia has maintained an aggressive one-year release cycle for new processors, diverging from the industry's typical two-year cadence. This pace, established in late 2023, has helped the company build a significant competitive advantage in the semiconductor market.
The AI sector has experienced a persistent shortage of chips with sufficient computational power for processing tasks. Nvidia's advanced chips have been in particularly short supply, prompting customers to explore alternative options from competitors.
Advanced Micro Devices recently announced its MI440 series data center GPUs, which will begin shipping early this year. Broadcom has gained attention as a more energy-efficient alternative through its application-specific integrated circuits.
Alphabet has developed custom Tensor Processing Units for internal use over the past decade and is reportedly considering leasing these chips to Meta Platforms. Amazon announced late last year that its latest Trainium chip offers 30% to 40% lower AI training costs compared to GPU-based alternatives.
Despite this competitive landscape, Nvidia's early production of Vera Rubin chips extends its market lead. The company also has clear visibility into future sales, with a backlog exceeding $500 billion as stated by Huang late last year. This backlog is expected to be filled over the six quarters ending in early 2027.
Since that backlog announcement, Nvidia has reported revenue of $57 billion and forecasts $65 billion for its upcoming fourth quarter. These figures suggest potential sales reaching approximately $378 billion next year, representing growth of 155%.
At an investor event last week, CFO Colette Kress noted that demand has increased since the $500 billion backlog estimate was provided. Kress stated the company would "definitely" surpass its previous outlook.
Nvidia stock experienced volatility last year, declining 37% from an all-time high in early 2025 before recovering. The stock has gained 977% over the past three years but currently trades 12% below its peak from early November.