Nvidia expects CPU revenue to more than double by fiscal year 2028, from roughly $20 billion in fiscal year 2027 to about $40 billion. CFO Colette Kress gave the fiscal 2027 figure during the company’s Q1 FY2027 earnings call, including standalone Grace and Vera CPUs and processors integrated into superchips. The strategy targets a $200 billion total addressable market for agentic AI workloads rather than traditional general-purpose computing. Early Vera demand is supporting deployments with a 1:1 CPU-to-GPU ratio, while CPUs currently contribute roughly 3% of Nvidia’s total revenue and analysts expect that share to reach approximately 5% by calendar year 2028. Wolfe Research projects Nvidia’s agentic CPU revenue at $6.6 billion in 2026, $14 billion in 2027 and $24.6 billion in 2028, compared with an x86 server CPU market of about $30 billion. Nvidia is using Arm architecture instead of x86, positioning the business as a parallel market to Intel and AMD’s established territory. Its broader proposition combines CPUs, GPUs, networking and software in one data-center stack, increasing competitive pressure on AMD, which sells EPYC server CPUs and competes with Nvidia in data-center GPUs. Investors will be watching whether the 1:1 CPU-to-GPU ratio persists as AI deployments expand.