Nvidia's H100 GPUs are going to space — Crusoe and Starcloud pioneer space-based solar-powered AI compute cloud data centers

AI factory company Crusoe and Starcloud are partnering to establish the first public cloud AI data centers in space, leveraging Nvidia H100 GPUs. This initiative, with initial H100 deployments by Starcloud in November 2025 and Crusoe Cloud services by early 2027, aims to utilize abundant solar power for significantly lower energy costs, estimated at 10x less than terrestrial operations. This development signals a new frontier for high-performance AI computing infrastructure, potentially disrupting traditional data center economics and energy sourcing for advanced AI workloads.

Analysis

Crusoe, an AI factory company, is partnering with Starcloud to establish the first public cloud AI data centers in space, utilizing Nvidia H100 GPUs. Starcloud, an Nvidia Inception program-backed entity, will launch the initial H100s in November 2025, with Crusoe Cloud services becoming available by early 2027. This collaboration aims to deploy "100x more powerful GPU (AI) compute" in orbit than previously achieved.

The strategic rationale centers on leveraging space-based solar power, which offers "almost unlimited, low-cost renewable energy" unencumbered by Earth's atmosphere. Estimates suggest these space-based data centers could achieve "10x lower energy costs (including launch cost)" compared to terrestrial operations. This energy-first approach also alleviates strain on Earth's energy grid and utilizes the vacuum of deep space for cooling.

This initiative represents a significant frontier in high-performance AI computing infrastructure, potentially disrupting traditional data center economics and energy sourcing. For Nvidia, the integration of its H100 GPUs into such an innovative and high-profile project underscores its continued dominance in AI hardware and expands its market reach into nascent, high-growth sectors. The positive sentiment surrounding this venture highlights investor optimism for novel AI infrastructure solutions.

Key Decisions for Investors

• Investors should monitor the progress of Crusoe and Starcloud's space-based data center deployment, particularly the successful integration and performance of Nvidia's H100 GPUs, as this represents a new, high-potential market for AI compute.
• Evaluate Nvidia's long-term strategic positioning in emerging, energy-efficient AI infrastructure, as this initiative could validate new applications and demand drivers for its advanced hardware.
• Consider the significant technological and operational risks associated with space-based infrastructure, including radiation hardening, orbital debris, and cooling efficacy, which could impact the projected cost savings and timelines.