The AI server CPU market is on track to expand to approximately $201 billion by 2030, driven by a projected 44% compound annual growth rate over the next five years, according to a new report from Raymond James.
The firm’s outlook aligns with Nvidia’s $200 billion long-term framework but trails AMD’s more aggressive $220 billion estimate, which could materialize under accelerated adoption scenarios. Raymond James’ analysis breaks down the 2030 market into three segments: conventional datacenter CPUs at $33.5 billion, AI head-end CPUs at $83 billion, and agentic CPUs at $85 billion. Agentic AI is identified as the fastest-growing segment, as persistent agents generate sustained workloads for retrieval, database management, and application execution primarily handled by CPUs.
Advanced Micro Devices received a rating upgrade to Strong Buy from Outperform, with analysts citing direct earnings leverage, strong datacenter positioning, and anticipated market-share gains. Raymond James also extended its financial model for AMD through 2028. The company’s second-quarter 2026 data center revenue surged 107% year-over-year, exceeding expectations and reinforcing its upward trajectory.
Nvidia, while maintaining a dominant position in AI accelerators, was noted for its rapidly growing CPU franchise, though still a smaller component relative to its core business. BMO Capital initiated coverage of Nvidia with an Outperform rating, projecting second-quarter revenues to significantly surpass consensus due to data center strength. KeyBanc similarly highlighted expectations for robust results, driven by the ramp-up of its Rubin R200 GPU shipments.
