GPU-as-a-service (GPUaaS): worldwide forecast 2025–2030

This report provides forecasts for service provider revenue from GPU-as-a-service (GPUaaS) offerings for 2025–2030. It provides details of revenue by service provider type, including hyperscaler cloud providers, alternative cloud providers and telecoms operators. Details are also provided for GPUaaS revenue by workload type and for sovereign cloud GPUaaS. The report offers recommendations for GPUaaS providers and the vendors that serve these providers.

This forecast report and accompanying data annex provide:

• a detailed worldwide revenue forecast for the GPUaaS market, split by:
  • provider type:
    • Hyperscaler cloud providers
    • Alternative cloud providers
    • Telecoms operators
  • GPUaaS revenue by workload type:
    • AI training versus AI fine-tuning versus AI inferencing versus non-AI 
  • Sovereign cloud GPUaaS revenue
• an examination of the market drivers and how they will affect spending during the forecast period
• analysis of the business environment and regional dynamics that will influence the GPUaaS market
• detailed market definitions
• recommendations for GPUaaS providers and AI hardware and software infrastructure vendors.

Questions answered in this report

• What is the current size of the GPUaaS market and how will it grow in the future?
• What are the main trends and factors that will affect the GPUaaS market during 2025–2030?
• How will GPUaaS spending be split between different categories of GPUaaS provider?
• How will the GPUaaS market evolve in different regions?
• What is the contribution of different types of workload to the GPUaaS market and how are the relative contributions changing over time?
• How will the sovereign cloud GPUaaS market evolve?

Geographical coverage