Google has signed a massive AI compute deal with SpaceX, agreeing to pay $920 million per month for access to around 110,000 NVIDIA GPUs and related infrastructure, as demand for Gemini Enterprise and AI workloads continues to surge.
TL;DR
- Google will pay SpaceX $920 million per month from October 2026 through June 2029.
- The deal covers around 110,000 NVIDIA GPUs, along with CPUs, memory, and related components.
- Google can terminate the agreement if SpaceX misses GPU delivery commitments.
- The deal strengthens SpaceX’s AI compute story ahead of its planned public market debut.
Google has entered into a major cloud services agreement with SpaceX, marking one of the biggest AI compute arrangements disclosed in the current infrastructure race.
According to a filing submitted to the U.S. Securities and Exchange Commission, SpaceX said it entered into a Cloud Service Agreement with Google LLC on June 5, 2026, covering access to compute capacity. The compute capacity includes around 110,000 NVIDIA GPUs, CPUs, memory, and other related components.
Under the terms of the agreement, Google will pay SpaceX $920 million per month from October 2026 through June 2029. SpaceX also noted that capacity will ramp up through September at a reduced fee before the full monthly structure takes effect.
The agreement includes safeguards for Google. If SpaceX fails to deliver access to the committed GPU capacity by September 30, 2026, then after a one-month grace period, Google can either terminate the agreement or accept the available number of GPUs with a pro-rated reduction in monthly fees.
The deal is also not locked beyond all exit options. After December 31, 2026, either party may terminate the agreement with 90 days’ notice. The filing also clarifies that Google will retain ownership and intellectual property rights over its content, AI models, and related data.
The reason behind the deal appears to be surging enterprise AI demand. A Google Cloud spokesperson, who was not named in the report, told Business Insider that “Google Cloud and SpaceX are long-time partners” and described the agreement as “bridge capacity” designed to meet rising customer demand for Gemini Enterprise, Google’s agentic AI platform.
The move comes as Google, Microsoft, Amazon, Meta, OpenAI, Anthropic, and other AI leaders continue to compete for scarce high-end compute capacity. For Google, the SpaceX agreement adds external AI infrastructure capacity at a time when customers are demanding more compute for AI agents, enterprise automation, model deployment, and productivity tools.
Topics For More Insights
For SpaceX, the agreement adds another high-value customer to its growing AI compute business just ahead of its expected U.S. stock-market debut. Reuters reported that SpaceX’s compute access deals with Google and Anthropic are worth around $26 billion annually combined, assuming the contracts continue as planned.
The Google deal follows SpaceX’s previously disclosed compute agreement with Anthropic. Reuters reported that Anthropic’s arrangement gives the Claude maker access to the full computing power of SpaceX’s Colossus 1 facility in Memphis, Tennessee, which houses more than 220,000 NVIDIA processors and adds 300 megawatts of new capacity.
TechCrunch, which first highlighted the Google agreement as another compute win for SpaceX ahead of its IPO, noted that the deal is similar in length and scope to the Anthropic arrangement.
The bigger takeaway is clear, AI compute is no longer just a back-end infrastructure cost. It has become a strategic asset, a revenue engine, and a market-positioning tool. Google gets near-term GPU capacity for Gemini Enterprise, while SpaceX strengthens its pitch as more than a rocket, satellite, and connectivity company.
