Nvidia’s new, more efficient racks could lessen the need for products that use cold water to remove excess heat from data centers.

The chip giant says Vera Rubin will sharply cut the cost of training and running AI models, strengthening the appeal of its integrated computing platform.

Wall Street also is hardly backing away from Nvidia, with 76 of the 82 analysts who cover the company holding buy ratings and only one recommending selling. The average Wall Street price target implies a gain of 37% over the next 12 months, which would push its market value over $6 trillion.

Nvidia finalized a $5B Intel stake after FTC approval, buying 214.7M shares. The move bolsters Intel’s foundry push as 18A ramps.

Nvidia CEO Jensen Huang said they can deliver five times the artificial-intelligence computing of the company’s previous chips when serving up chatbots and other AI apps.

Herbert Ong, an investor who has tracked Tesla and Nvidia for years, translated that into a simple investor thesis. Nvidia is “starting at the platform level” and “staying at the platform level,” while Tesla is building “the full AI stack” into its own cars, Ong said in a recent discussion of the CES announcements on his YouTube channel.

Nvidia unveiled the Vera Rubin AI computing platform at CES 2026, claiming up to 10x lower inference token costs and faster training for MoE models.

Equity research firm Evercore ISI said recently that it sees Nvidia stock at $352 by the end of 2026. That would be an increase of around 86% from the current stock price. Analyst Mark Lipacis raised his price target on the stock from the earlier level of $261, stating that Nvidia's revenue growth could jump to 79% by the middle of 2026.