SAN FRANCISCO, Feb 17 (GeokHub) — Nvidia has signed a multiyear agreement to supply Meta with millions of artificial intelligence chips, deepening one of the semiconductor industry’s most significant partnerships.
The deal covers both Nvidia’s current Blackwell AI processors and its next-generation Rubin chips, as well as standalone installations of its Grace and upcoming Vera central processing units (CPUs). Financial terms were not disclosed.
Nvidia’s Grace and Vera CPUs are designed to complement its powerful AI accelerators but are increasingly positioned as standalone data center processors. The company introduced Grace in 2023, leveraging technology from Arm Holdings.
The latest agreement signals Nvidia’s intention to compete more directly with established CPU makers such as Intel and Advanced Micro Devices.
According to Nvidia executives, Grace processors have demonstrated significant energy efficiency improvements for workloads such as running large-scale databases, with the next-generation Vera chip expected to deliver even greater performance gains.
The company said Meta has already tested Vera processors in select workloads, with early results described as promising.
Strategic Implications
The partnership comes at a time when Meta is actively developing its own AI silicon and exploring additional hardware collaborations. The social media giant has also held discussions with Google regarding the potential use of its Tensor Processing Units (TPUs) for AI workloads.
Despite these efforts, Nvidia remains a central supplier to hyperscale cloud and AI operators. Industry analysts widely believe Meta is among Nvidia’s largest customers, contributing significantly to its recent revenue growth.
By highlighting the multiyear agreement, Nvidia appears to be reinforcing its position not only as the dominant AI accelerator provider but also as a serious contender in the broader data center CPU market.
The deal underscores the escalating competition among technology giants to secure long-term access to advanced AI hardware. As generative AI models grow larger and more computationally demanding, reliable supply of high-performance chips has become a strategic priority.
For Nvidia, expanding CPU adoption alongside its AI accelerators strengthens its integrated data center strategy — positioning the company at the center of the global AI infrastructure buildout.
