Nvidia and AMD are committing a combined $4 billion to Meta as both chipmakers work to lock in long-term AI infrastructure business with one of the largest hyperscalers. The investments come as Meta ramps up spending on AI compute and expands its data center footprint.
Meta disclosed on Monday, October 26, that it plans to raise $97.5 billion through an equity offering to fund continued AI infrastructure investment. Nvidia announced a $3 billion investment in Meta the same day, with AMD following a day later with a $1 billion commitment.
Nvidia’s Expanded Role
As part of its deepened partnership with Meta, Nvidia said the two companies plan to deploy more of its infrastructure, including the Vera Rubin platform, Vera CPUs for next-generation AI agents, Spectrum-X Ethernet networking, and BlueField-4 DPUs. Nvidia and Meta also said they will deepen collaboration on Meta’s Muse models and agents, which run on Nvidia infrastructure. The companies are also weighing additional work involving Nvidia’s OpenShell Reference Architecture to strengthen security for AI agents.
AMD Highlights Early Deployments
AMD used its announcement to point to concrete progress in its own Meta partnership. The company said initial AMD Helios rack-scale systems have already been installed and are being tested by Meta ahead of a wider rollout. AMD also said Meta has deployed millions of Epyc processors and is a lead customer for its sixth-generation Epyc CPUs, which according to AMD support the dedicated computing environments behind Meta’s Muse personal AI agent.
Meta is also using AMD Instinct GPUs for training, ranking, and recommendation workloads, and plans to deploy a custom AMD Instinct GPU built specifically for its own workloads, AMD said.
Why Chipmakers Are Betting Big
Analysts framed the investments as part of a broader strategy by both chipmakers to secure future business with major hyperscale customers. Matt Kimball, vice president and principal analyst at Moor Insights & Strategy, said the moves continue a pattern of both companies investing in strategic hyperscalers to drive adoption of their platforms.
Brendan Burke, research director at Futurum Group, said Nvidia and AMD both view Meta’s AI agent strategy as among the more promising in the industry. He noted that Meta’s recent progress in model training and its personal AI agent work suggest the company has become a viable frontier AI lab, which explains its rising compute spending. That gives both chipmakers incentive to support Meta’s roadmap more aggressively, with the payoff expected through future GPU and CPU purchases.