AMD (AMD) Partnership summary

Opening remarks and agenda

• Conference call announced to discuss an expanded multi-year, multi-generation AI partnership, with executive leadership present and forward-looking statements highlighted.

• Announcement of up to 6 gigawatts of Instinct GPUs for AI infrastructure at Meta, with significance for scaling AI and diversifying compute resources.

Objectives of the partnership

• Multi-year, multi-generation agreement positions the company at the core of Meta's next-gen AI infrastructure and accelerates AI model development and deployment.

• Partnership aims to deliver high-performance, energy-efficient compute and AI platforms optimized for Meta's workloads.

• Alignment of silicon, systems, and software roadmaps for purpose-built AI solutions.

Partner introductions and roles

• Meta to be a lead customer for 6th Gen EPYC CPUs, codenamed “Venice” and “Verano,” with workload-specific optimizations.

• Meta has been a close, multi-generation partner, deploying millions of CPUs and hundreds of megawatts of GPUs globally.

• AMD to provide custom Instinct GPUs and EPYC CPUs, with Meta deploying these at global scale.