Advanced Micro Devices (AMD) and Meta have signed a multi-year, multi-generation agreement to deploy up to 6 gigawatts of AMD Instinct GPUs to power Meta’s AI infrastructure, the companies said Tuesday.
The first deployment will support 1 gigawatt of capacity, with shipments scheduled to begin in the second half of 2026. The systems will use a custom AMD Instinct GPU based on the MI450 architecture and 6th Gen AMD EPYC CPUs, codenamed “Venice,” running ROCm software.
The infrastructure will be built on AMD’s Helios rack-scale architecture, developed jointly with Meta through the Open Compute Project.
The agreement expands an existing partnership and aligns the companies’ GPU and CPU silicon, systems and software roadmaps. Meta has deployed AMD EPYC CPUs and Instinct MI300 and MI350 series GPUs across its global infrastructure over multiple generations.
“We are proud to expand our strategic partnership with Meta as they push the boundaries of AI at unprecedented scale,” said Lisa Su, CEO of AMD. “This multi-year, multi-generation collaboration across Instinct GPUs, EPYC CPUs and rack-scale AI systems aligns our roadmaps to deliver high-performance, energy-efficient infrastructure optimised for Meta’s workloads, accelerating one of the industry’s largest AI deployments and placing AMD at the centre of the global AI buildout.”
Mark Zuckerberg, founder and CEO of Meta, said the deal supports the company’s efforts to diversify its compute base.
“We’re excited to form a long-term partnership with AMD to deploy efficient inference compute and deliver personal superintelligence,” Zuckerberg said. “This is an important step for Meta as we diversify our compute. I expect AMD to be an important partner for many years to come.”
In addition to GPUs, Meta will be a lead customer for AMD’s 6th Gen EPYC CPUs, including “Venice” and “Verano,” a next-generation EPYC processor designed with workload-specific optimisations. AMD said CPUs remain a key part of AI infrastructure, supporting efficiency and orchestration alongside GPUs.
As part of the agreement, AMD has issued Meta a performance-based warrant for up to 160 million shares of AMD common stock. The warrant will vest in tranches tied to shipment milestones, beginning with the initial 1 gigawatt and scaling to 6 gigawatts. Vesting is also linked to stock price thresholds and technical and commercial milestones.
“We expect this partnership to drive substantial multi-year revenue growth and be accretive to our non-GAAP earnings per share, marking another significant step forward in delivering on our ambitious long-term financial model,” said Jean Hu, AMD’s EVP, CFO and treasurer.
The companies said they will collaborate across silicon, systems and software to scale AI infrastructure globally.
ALSO READ: Fei-Fei Li’s World Labs Raises $1 Billion to Scale Spatial AI
