Meta ramps up Nvidia: Deploying millions of chips in the coming years, first time adopting Grace CPU

```

Meta has agreed to deploy “millions” of Nvidia chips over the coming years, further strengthening the already close collaboration between these two AI industry giants.

According to a statement released on Tuesday, Meta has committed to using more AI processors and networking equipment from Nvidia. At the same time, Meta will for the first time adopt Nvidia’s Grace CPU as a core component in its standalone computers. This deployment will cover Nvidia’s current Blackwell architecture as well as the upcoming Vera Rubin-designed AI accelerator products.

Meta contributes 9% of Nvidia’s revenue. Meta CEO Mark Zuckerberg said in the statement: “We are excited to expand our collaboration with Nvidia, using their Vera Rubin platform to build state-of-the-art compute clusters and deliver personal superintelligence to everyone in the world.”

This agreement, announced amid ongoing shifts in the AI competitive landscape, reaffirms Meta’s loyalty to Nvidia. Currently, Nvidia’s systems are still regarded as the gold standard for AI infrastructure, generating hundreds of billions of dollars in revenue for the company. However, competitors are offering alternatives, and Meta is also developing its own chip components.

After the news was released, the share prices of both Nvidia and Meta rose by more than 1% in after-hours trading, while Nvidia’s rival AMD fell over 3%.

Ian Buck, Nvidia’s vice president of accelerated computing, said:

The two parties did not disclose the specific investment amount or timetable. He noted that it is reasonable for companies like Meta to test other alternatives, but he emphasized that only Nvidia can provide the complete components, systems, and software ecosystem that a company hoping to stay ahead in AI needs.

Meanwhile, Zuckerberg has made AI Meta’s top priority, pledging to invest hundreds of billions of dollars in building the next generation of infrastructure:

Meta expects to set a new spending record in 2026. Zuckerberg said last year that the company will invest $60 billion in U.S. infrastructure projects over the next three years. Meta is building several gigawatt-scale data centers across the U.S., including in Louisiana, Ohio, and Indiana. One gigawatt of power is enough to supply about 750,000 homes.

Buck emphasized that Meta will be the first major data center operator to use Nvidia CPUs in standalone servers. Typically, Nvidia offers such CPUs together with its high-end AI accelerators.

This move means Nvidia is further entering an area traditionally dominated by Intel and AMD. At the same time, it offers an alternative for large data center operators developing their own chips, such as those designed by Amazon AWS.

Buck said that the application scenarios for such chips are continually growing:

As the parent company of Facebook and Instagram, Meta will not only use these chips itself but also adopt computing power provided by other companies based on Nvidia’s architecture.

Nvidia CPUs will increasingly be used for tasks such as data processing and machine learning.

CPUs have many different types of workloads. We have found that Grace is an excellent back-end data center CPU, meaning it is very good at handling background computational tasks. On these back-end workloads, it can actually deliver double the performance per watt.

Risk Warning and DisclaimerThe market has risks, and investment needs to be cautious. This article does not constitute personal investment advice and has not taken into consideration the special investment objectives, financial situation or needs of any individual user. Users should consider whether any opinions, views, or conclusions in this article are suitable for their particular situation. Investing accordingly is at your own risk. ```