Austin, Texas, USA, July 6, 2026

Futurum’s new five-year forecast finds the global data center semiconductor market reaching $1.2 trillion by 2030 as agentic AI inference reshapes silicon demand.

The global data center semiconductor market will grow more than fivefold to $1.2 trillion by 2030, driven by a decisive shift from AI training to high-concurrency agentic inference, according to a new report from The Futurum Group.

The global data center semiconductor market reached $241.0 billion in CY2025 (see Figure 1), up 3.6x from $66.6 billion in CY2023, and Futurum Research’s base case carries it to $1,212.7 billion by CY2030 at a 38.1% five-year CAGR. Growth is front-loaded — more than doubling in CY2026 and rising 54.0% in CY2027 as committed capacity lands — before decelerating as the market matures. A bull case reaches $2.2 trillion on faster continuous-learning investment, while a bear case compresses to $746.6 billion by CY2030 on ROI scrutiny, packaging bottlenecks, and grid constraints.

The forecast’s organizing thesis is that agentic inference, not training, now sets semiconductor roadmaps. Inference-focused accelerators will grow from roughly half the CY2025 market to 73% by CY2030 ($884.9 billion), and Agent and Reasoning-first inference alone reaches $546.0 billion — 45% of all data center semiconductor revenue.

Figure 1: Data Center Semiconductors: Base Case Forecast by Segment, 2025-2030

“2026 is the first year of the inference transition that defines the rest of the decade,” said Brendan Burke, Research Director, Semiconductors, Supply Chain & Emerging Tech at Futurum. “High-concurrency agentic workloads will set the architecture for vendor roadmaps through 2027.”

The research reveals several key developments reshaping the data center semiconductor landscape:

• GPUs hold half the market at $604.5B by 2030 (30.8% CAGR), but custom XPUs grow fastest at a 44.6% CAGR to $237.2B and run roughly 75% cheaper than equivalent GPU deployments.
• Off-chip memory is the single fastest-growing line at a 72.4% CAGR to $260.5B, while HBM grows 5.9x to a complementary $197.6B market as bandwidth becomes the binding constraint on agentic inference.
• CPUs stage an unexpected renaissance, from $28.9B to $110.4B, as agentic orchestration pushes CPU-to-GPU ratios back toward 1:1 and beyond.
• Hyperscalers remain the largest buyer at $606.5B by 2030 (50% of the market), while Tier 2 cloud is the fastest-growing deployment model at a 52.5% CAGR.
• The United States remains the largest region at $477.7B by 2030 (39% share), while Asia ex-China grows fastest at a 45.6% CAGR.

“At a 75% cost advantage, every inference workload that can run on an XPU eventually will, and the binding constraint is advanced packaging throughput,” noted Burke. “On memory, the supercycle is real, but buyers should build flexibility into every commitment before the cost structure resets in 2028.”

Read more in the report “1H 2026 Data Center Semiconductors Market Sizing & Five-Year Forecast” on the Futurum Intelligence Platform.

About Futurum Intelligence for Market Leaders

Futurum Intelligence’s Semiconductors IQ service provides actionable insight from analysts, reports, and interactive visualization datasets, helping leaders drive their organizations through transformation and business growth. Subscribers can log into the platform at https://app.futurumgroup.com/, and non-subscribers can find additional information at Futurum Intelligence.

Follow news and updates from Futurum on X and LinkedIn using #Futurum. Visit the Futurum Newsroom for more information and insights.

Other Insights From Futurum:

Futurum Signal Report | AI Cloud Platforms

1H 2026 Data Center Semiconductor Decision Maker Survey Report – Subscribers

Futurum Signal Report I AI Accelerators