In another wild turn for AI chips, Meta signs deal for millions of Amazon AI CPUs

2026-04-24 · Source: AI News & Artificial Intelligence | TechCrunch · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Cloud Computing & IT Infrastructure, Robotics & Autonomous Systems · Depth: Novice, quick

Summary

Meta has signed a deal with Amazon Web Services (AWS) to utilize millions of AWS Graviton chips for its expanding artificial intelligence (AI) needs. The Graviton is an ARM-based CPU designed to handle compute-intensive AI workloads such as real-time reasoning, code generation, search, and multi-step agent coordination, which are distinct from the GPU-centric demands of large model training. This agreement follows Meta's prior $10 billion, six-year deal with Google Cloud and highlights the intensifying competition among cloud providers, particularly as Google also develops its own custom AI chips. AWS additionally offers its Trainium AI GPU, which Anthropic has committed to using extensively as part of a $100 billion cloud spending agreement over 10 years, with Amazon investing $13 billion into Anthropic.

Key takeaway

For AI architects evaluating infrastructure for agentic workloads, your decision should now heavily weigh AWS Graviton CPUs. This deal signals a strong validation for ARM-based CPUs in AI inference and real-time reasoning tasks, potentially offering better price-performance ratios than traditional GPUs for specific post-training applications. Explore AWS Graviton 5 for your next-generation AI agent deployments.

Key insights

AWS secured a significant Meta deal for Graviton CPUs, intensifying cloud and custom AI chip competition.

Principles

• AI agent workloads shift chip requirements.
• Cloud providers prioritize custom silicon.
• Price-performance drives enterprise AI adoption.

In practice

• Consider ARM-based CPUs for AI inference.
• Evaluate cloud provider custom chips for cost.
• Diversify cloud infrastructure for AI workloads.

Topics

• AWS Graviton
• AI CPUs
• AI Agentic Workloads
• Cloud Computing Competition
• Amazon Trainium

Best for: VP of Engineering/Data, Investor, AI Architect, AI Hardware Engineer, Director of AI/ML, CTO

Related on AIssential

• Meta buys tens of millions of AWS Graviton 5 processor cores from Amazon — Meta is heavily investing in ARM-based CPUs, including AWS Graviton 5, to power agentic AI systems.
• Amazon’s Newest Gambit: Selling AI Chips — Amazon and Google are selling custom AI chips, challenging Nvidia's market dominance in specific workloads.
• Alibaba Launches XuanTie C950 CPU for Agentic AI — Open-standard RISC-V architecture enables competitive, sovereign AI hardware development.
• Anthropic and Amazon expand collaboration for up to 5 gigawatts of new compute — Anthropic secures massive AWS compute and investment to scale Claude's training and inference capabilities.
• Nvidia and Meta Agree to Wide-Ranging new AI Chip Deal — Meta and Nvidia deepen their AI hardware partnership, focusing on next-gen GPUs, CPUs, and networking for large-scale AI infrastructure.
• An exclusive tour of Amazon’s Trainium lab, the chip that’s won over Anthropic, OpenAI, even Apple — Amazon Web Services (AWS) is significantly advancing its custom AI chip development, underscored by a \$50 billion investment deal with OpenAI that includes supplying 2 gigawatts of Trainium computing capacity.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by AI News & Artificial Intelligence | TechCrunch.