AMD and Meta strike a landmark 100 billion dollar, 6-gigawatt chip procurement and power supply agreement.

AMD has secured a multiyear agreement to supply up to 6 GW of AI capacity to Meta, matching a similar deal AMD struck with OpenAI in October 2025. Deployments will begin later this year using a custom AMD Instinct GPU based on the MI450 architecture in AMD's Helios rack-scale servers, paired with AI-optimized AMD EPYC CPUs. Meta plans to use AMD's chips primarily to run inference for its AI models, stating in a press release: "We need massive, scalable compute power that can handle the growing demands of our AI workloads. Our partnership with AMD, which builds on our existing collaboration, will help us meet those needs."

As part of the agreement, AMD issued Meta a performance-based warrant for up to 160 million AMD shares—potentially about a 10% stake—structured to vest in tranches based on hitting milestones up to 6 GW. Vesting is also tied to AMD's share-price thresholds, up to $600 per share, and the exercise requires Meta to meet technical and commercial milestones. Following the announcement, Nvidia's shares were down slightly with the chipmaker scheduled to report fourth-quarter results on February 25, with analysts expecting data center sales at around $61 billion, while AMD's stock price jumped more than 9%.

AMD CEO Lisa Su said in a statement: "This multiyear, multi-generation collaboration across Instinct GPUs, EPYC CPUs, and rack-scale AI systems aligns our roadmaps to deliver high-performance, energy-efficient infrastructure optimized for Meta's workloads, accelerating one of the industry's largest AI deployments and placing AMD at the center of the global AI buildout." Meta CEO Mark Zuckerberg expressed confidence in the relationship, stating: "We're excited to form a long-term partnership with AMD to deploy efficient inference compute and deliver personal superintelligence. ... I expect AMD to be an important partner for many years to come."

The deal represents significant validation for AMD as the industry shifts toward inference economics, with Nvidia still commanding large-scale training for frontier AI models. This comes just a week after Meta announced a long-term infrastructure deal with Nvidia for hardware, including GPUs and Spectrum-X Ethernet networking platform. According to Matt Kimball, vice president and principal analyst at Moor Insights & Strategy, the AMD agreement demonstrates both scale and strategic importance: "This is significant for AMD for a few reasons: massive volume commit, architectural integration into Meta's AI roadmap, and strategic credibility. These are all big deals, and credibility is perhaps the most important as the company looks to establish a more significant footprint."