무슨 발표인가

• 다수 세대 커스텀 MTIA 칩 협력 확대
• 2년 내 4세대 개발·배포 계획
• 추론·추천 및 생성AI 워크로드 지원

향후 2년 내 4개 세대

원문 (영어)

Today we’re announcing an expanded partnership with Broadcom to co-develop multiple generations of our next-generation MTIA (Meta Training and Inference Accelerator) chips — the custom silicon that helps power AI across all of Meta’s apps and services.

At Meta, we take a portfolio approach to AI silicon, matching the right accelerator to each workload to achieve the best mix of performance and total cost of ownership. MTIA is our purpose-built accelerator optimized for inference and recommendation at scale — and this partnership with Broadcom will help us deliver it at the level our ambitions require.

We recently announced that we’re developing and deploying four new generations of MTIA chips within the next two years to support ranking and recommendations, along with generative AI workloads, and this agreement will support the acceleration of our custom silicon.

Broadcom will work with us across chip design, advanced packaging, and networking to help build out the massive computing foundation we need to deliver real-time AI experiences to billions of people. The partnership is built on Broadcom’s XPU platform, a technology designed for creating custom AI accelerators (XPUs), enabling optimization of Meta’s AI infrastructure across multiple silicon generations.

Broadcom’s advanced Ethernet technologies will also enable seamless, high-bandwidth networking across Meta’s rapidly expanding AI compute clusters. The agreement includes a commitment that exceeds 1GW, which is the first phase of a sustained, multi-gigawatt rollout, reinforcing a shared roadmap to co-design and scale the hardware required to bring personal superintelligence to billions of people globally.

원문: Meta Newsroom — "Meta Partners With Broadcom to Co-Develop Custom AI Silicon" (2026-04-14) 공식 원문: https://about.fb.com/news/2026/04/meta-partners-with-broadcom-to-co-develop-custom-ai-silicon/