Starcloud's Orbital AI Data Centers: Nvidia H100 Powers the First GPU in Space as $170M Funding Fuels Expansion

Starcloud's breakthrough deployment of the first Nvidia H100 GPU in space represents a pivotal advancement in orbital AI computing, addressing terrestrial limitations on energy, land, and cooling while capitalizing on Nvidia's dominant AI hardware ecosystem. This development is commercially significant as it opens new revenue streams for Nvidia through space-qualified GPUs, with Starcloud's rapid unicorn status validating the market potential. North American investors should note the alignment with U.S.-based innovation in low-Earth orbit infrastructure, potentially amplifying Nvidia's growth beyond Earth-bound data centers.

As of: 30.03.2026

By Dr. Elena Voss, AI Infrastructure Analyst: Starcloud's space AI push exemplifies how Nvidia's GPUs are transcending planetary boundaries, reshaping compute markets for the orbital era.

Starcloud Launches First Nvidia H100 GPU into Orbit

Starcloud marked a historic achievement in November 2025 by launching Starcloud-1, a 60kg satellite carrying an Nvidia H100 GPU, making it the most powerful GPU compute platform ever operated in space by a factor of approximately 100.

This deployment enabled the first AI model training in orbit and ran a version of Google's Gemini model, demonstrating practical AI inference and training capabilities in low-Earth orbit.

The H100, Nvidia's flagship data center GPU, powers complex AI workloads on Earth; its space adaptation highlights Starcloud's engineering prowess in radiation-hardening and thermal management for vacuum environments.

Starcloud's CEO Philip Johnston emphasized the satellite's role in proving orbital compute viability, setting the stage for scalable data centers in space.

This positions Starcloud ahead of competitors, as no other entity has achieved comparable GPU performance in orbit.

$170 Million Series A at $1.1 Billion Valuation

Just 17 months after its Y Combinator demo day, Starcloud secured $170 million in Series A funding led by Benchmark and EQT Ventures, achieving unicorn status faster than any prior YC company.

The capital will fund Starcloud-2, slated for October 2026 launch, featuring multiple GPUs including a Nvidia Blackwell chip, an AWS server blade, and a bitcoin mining computer, along with the largest deployable radiator ever on a private satellite.

Starcloud-2 will generate 100 times more power than its predecessor and host commercial cloud workloads for customers like Crusoe, marking the transition from proof-of-concept to revenue-generating operations.

Further plans include Starcloud-3, a 200-kilowatt, three-tonne spacecraft compatible with SpaceX's Starship deployment system, aiming for cost-competitiveness with ground-based facilities.

This funding underscores investor confidence in space-based AI, bypassing Earth constraints like energy costs and land scarcity.

Nvidia's Rubin Architecture and Broader AI Momentum

Parallel to Starcloud's advances, Nvidia is transitioning to its Rubin (R200) architecture, unveiled at CES 2026, utilizing TSMC's 3nm process and HBM4 memory for 3x efficiency gains in massive AI models over Blackwell.

The accompanying Vera CPU, an 88-core design, pairs with Rubin GPUs to minimize reliance on third-party processors, enhancing integrated systems for data centers—including potential orbital ones.

CEO Jensen Huang projects $1 trillion in sales from Blackwell and Rubin platforms by 2027, fueled by demand from cloud providers and enterprises.

Omdia reports $110.9 billion in Q4 2025 cloud infrastructure spend, up 29% YoY, with 27% growth projected for 2026, directly boosting Nvidia GPU demand.

Strategic Relevance for Commercial AI Compute

Orbital data centers like Starcloud's address key bottlenecks: unlimited solar power, zero land costs, and radiative cooling superior to terrestrial systems.

By hosting Nvidia GPUs, Starcloud enables hyperscalers to scale AI without grid strain, relevant as AI training demands explode with models like Mixture-of-Experts (MoE).

Partnerships, such as with Crusoe for workloads, signal enterprise adoption, while Starcloud's U.S. base in Redmond, Washington, aligns with North American tech hubs.

Competitors like Rebellions target niche AI inferencing with NPUs, but Starcloud leverages Nvidia's GPU dominance for broad applicability.

Investor Context Amid JD.com Ecosystem

While Starcloud operates independently, its Nvidia-centric model ties into broader supply chains influencing investors in JD.com (ISIN: KYG5635P1090), whose online retail platform increasingly integrates AI for logistics and personalization.

JD.com's global.jd.com portal showcases AI-driven e-commerce innovations, potentially benefiting from orbital compute efficiencies in the long term.

North American investors tracking Chinese tech exposure via JD.com shares should monitor space AI as a diversifying growth vector, though direct links remain speculative.

IR updates at corporate.jd.com provide filings on strategic tech investments.

Regulatory and Geopolitical Landscape

The Chip Security Act of 2026 tightens export controls, yet Nvidia resumed limited H200 sales to China under volume caps, maintaining revenue streams.

EU and U.S. antitrust scrutiny on Nvidia's AI stack dominance could impact partnerships, but space applications may evade some terrestrial regulations.

Starcloud's domestic U.S. operations mitigate geopolitical risks associated with orbital tech.

Future Outlook for Orbital AI

With Starcloud-2 and -3 on horizon, orbital AI could capture a slice of the $1 trillion AI infrastructure market by 2030.

Nvidia's roadmap, including Rubin, positions it as the enabler, with data center revenue already at 85%+ of total.

For North American investors, this convergence of space tech and AI hardware offers exposure to high-growth, frontier markets.

Disclaimer: Not investment advice. Stocks are volatile financial instruments.