Meta announced on March 11, 2026 that it has introduced four new custom AI chips—MTIA 300, MTIA 400, MTIA 450, and MTIA 500—designed to power its expanding data‑center AI workloads and reduce dependence on external vendors such as Nvidia and AMD.
The MTIA 300, already deployed a few weeks earlier, is focused on training smaller AI models that underpin Meta’s ranking and recommendation systems across Facebook, Instagram, WhatsApp, and Threads. The MTIA 400 has completed testing and will be deployed in Meta’s data centers in the near term, while the MTIA 450 and MTIA 500 are targeted for generative‑AI inference tasks, including image and video creation, and are expected to become operational by the end of 2027. The MTIA 450 doubles the high‑bandwidth memory (HBM) bandwidth of the MTIA 400, and the MTIA 500 increases it by an additional 50%, delivering the performance needed for large‑scale inference workloads.
Meta’s custom silicon strategy is part of a broader plan to front‑load AI infrastructure. The company is investing $70‑$72 billion in capital expenditures for 2025 and $115‑$135 billion for 2026, a significant portion of which is earmarked for data‑center expansion, server procurement, and chip development. By building its own silicon, Meta aims to lower cost per performance, accelerate deployment of AI workloads, and maintain control over its hardware roadmap.
The rapid release cadence—new chip generations every six months—positions Meta ahead of competitors such as Google’s TPUs and Amazon’s Trainium and Inferentia chips. The inference‑first design of the MTIA 450 and MTIA 500 aligns with Meta’s focus on operational cost efficiency, as inference represents a growing share of AI compute expenses. The partnership with Broadcom for design elements and TSMC for fabrication further diversifies Meta’s supply chain and mitigates vendor risk.
These developments signal a strategic pivot toward vertical integration in AI infrastructure, potentially reshaping Meta’s competitive dynamics and cost structure in the long term.
The content on EveryTicker is for informational purposes only and should not be construed as financial or investment advice. We are not financial advisors. Consult with a qualified professional before making any investment decisions. Any actions you take based on information from this site are solely at your own risk.