Nvidia CEO Jensen Huang used his keynote speech at the Consumer Electronics Show (CES) in Las Vegas to confirm that the company’s forthcoming Vera Rubin chips are in “full production.”

First unveiled at Nvidia’s GTC event in March 2025, the Vera Rubin Superchip, which consists of an Arm-based Vera CPU and a Rubin GPU, is expected to be available in the second half of 2026.

The Rubin platform comprises six chips in total – the NVLink 6 Switch, ConnectX-9 SuperNIC, BlueField-4 DPU, and Spectrum-6 Ethernet Switch, in addition to the Vera CPU and Rubin GPU.

Vera is the successor to Nvidia’s Grace CPU, while Rubin will succeed Blackwell, with Nvidia claiming Rubin will be capable of achieving 5x and 3.5x of inference and training performance, respectively, compared to Blackwell.

The Vera Rubin NVL72 rack will also differ to previous generations in that it will be 100 percent liquid cooled and feature cable‑free modular tray designs, which Nvidia claims will allow installation times to be reduced from two hours to five minutes, in comparison to its Blackwell equivalent.

Rubin will be followed by the Rubin Ultra in the second half of 2027. All six Rubin chips had previously been unveiled by Nvidia throughout 2025.

“Rubin arrives at exactly the right moment, as AI computing demand for both training and inference is going through the roof,” said Huang. “With our annual cadence of delivering a new generation of AI supercomputers — and extreme codesign across six new chips — Rubin takes a giant leap toward the next frontier of AI.”

Huang’s speech comes almost two weeks after it was announced that Nvidia plans to license technology from AI chip startup Groq and hire its leadership team. Though the deal is not an outright acquisition, it is reportedly worth up to $20 billion.

No mention of the agreement was made by Huang during his speech, but according to an analyst at Citrini, Nvidia was questioned about the Groq licensing agreement during the CES Financial Analyst Q&A.

An analyst from the Bank of America reportedly asked about the short, medium, and long-term significance of the Groq licensing deal, enquiring if it meant the industry has reached the stage where it’s believed that more "application-specific integrated circuits (ASIC)-like" chips specialized for certain types of inference are needed?"

In response, Huang said that Nvidia’s chips are “inherently ASICs,” as they are designed in-house and manufactured in TSMC foundries.

“What Groq pursues is extreme low latency, which is in a trade-off relationship with our strength in extreme high throughput. In the future, when AI-powered devices (such as glasses) emerge, diverse demand for low latency will arise.”

Speaking at an event in October 2025, Google’s VP and GM of AI and infrastructure, Amin Vahdat, said he believes the future lies in specialization, noting that for some computations, “if you built even more specialized systems for… for example, maybe serving agentic workloads, that would benefit from an even more specialized architecture.”