Jensen Huang announced on Monday at the Consumer Electronics Show in Las Vegas that all six chips required for the next-generation computing platform named after astronomer Vera Rubin have completed production from manufacturing partners, with customer deployments expected in the second half of this year.

"Market demand is extremely robust," he stated, noting that the increasing complexity and proliferation of AI software applications are straining existing compute resources, driving the need for more powerful capabilities.

The Santa Clara, California-based company is striving to maintain its leadership position as the premier manufacturer of AI accelerators—core chips used by data center operators to develop and run AI models.

Some Wall Street analysts express concerns over intensifying competition for NVIDIA and potential difficulties in sustaining current AI spending growth rates. Data center operators are also developing in-house AI accelerators, though NVIDIA maintains a long-term market outlook valued in the trillions of dollars.

According to the company, Rubin represents the latest generation of accelerators, delivering 3.5x improvement in AI training performance and 5x better operational efficiency compared to its predecessor, Blackwell. The new central processor features 88 core computing units, doubling the performance of the model it replaces.

Compared to previous years, NVIDIA revealed product details earlier this year to sustain industry demand for its hardware—the technological foundation enabling the AI application boom. The company typically discloses product specifics at its spring GTC conference in San Jose, California.

While promoting new offerings, NVIDIA emphasized the continued strong performance of previous-generation products. The company also noted significant Chinese demand for the H200 chip, which had been included on the U.S. government's list of products potentially approved for export to China.

Chief Financial Officer Colette Kress informed analysts that license applications have been submitted and are under government review. She emphasized that regardless of approval levels, NVIDIA can maintain Chinese customer supply without disrupting global delivery capabilities.

For Huang, this CES appearance marks the latest stop in his series of promotional activities to drive AI system deployment. Key competitor AMD's CEO Lisa Su is also scheduled to deliver a keynote at the event.

The new product lineup includes networking and connectivity components, forming part of the DGX SuperPod supercomputer while also supporting modular standalone purchases by customers. The performance upgrade stems from AI technology's shift toward more specialized model networks that process massive input data and solve specific problems through multi-stage processes.

The company particularly highlighted that Rubin-based systems will have lower operational costs than Blackwell versions due to achieving equivalent performance with fewer components. Major remote computing providers like Microsoft are expected to lead hardware deployments in the second half of the year.

Currently, NVIDIA's computer system purchases are primarily funded through capital expenditure budgets of a few key customers including Microsoft, Google Cloud, and Amazon AWS. The company is driving AI adoption across economic sectors like robotics, healthcare, and heavy industry through coordinated hardware-software innovations.

As a strategic initiative, NVIDIA simultaneously released a toolkit suite to accelerate autonomous vehicle and robotics development.

The NVIDIA Alpamayo suite develops open AI models, simulation tools, and datasets designed to expedite next-generation safe, reasoning-based autonomous vehicle (AV) development.

Alpamayo introduces reasoning-based Chain-of-Thought visual-language-action (VLA) models that incorporate human-like reasoning into autonomous vehicles' decision-making processes.

Supported by NVIDIA Halos safety systems, these models enable step-by-step reasoning through novel scenarios, enhancing driving capabilities and explainability—critical for building trust and safety in intelligent vehicles.

Jensen Huang stated, "The ChatGPT moment for physical AI has arrived—machines are beginning to understand, reason, and act in the real world," said Jensen Huang, founder and CEO of NVIDIA. "Robo-taxis are among the first beneficiaries."

References:

https://nvidianews.nvidia.com/news/alpamayo-autonomous-vehicle-development

https://www.bloomberg.com/news/articles/2026-01-05/nvidia-ceo-says-new-rubin-chips-are-on-track-helping-speed-ai