SpaceX has secured a binding commitment for roughly $40 billion in financing to pay for AI chips and related data center equipment, according to a regulatory filing made public on Monday, October 26, 2026. The company described the arrangement as the largest committed GPU financing ever put together, and it marks one of the biggest single bets on artificial intelligence hardware by any private company to date.
The deal is structured as a lease rather than a straightforward loan. An entity managed by Apollo Global Management, the New York based private equity and credit giant, will buy the graphics processing units and the server hardware that goes with them, then lease that equipment back to SpaceX as it is delivered.
How the Financing Is Structured
Under the terms disclosed in the filing, each tranche of equipment will be leased to SpaceX for approximately five to five and a quarter years, with the clock starting at the point of delivery rather than at signing. That structure spreads the cost of the hardware across its useful life and keeps a large portion of the borrowing off SpaceX's own balance sheet, since a special purpose vehicle holds the physical assets.
SpaceX expects the transaction to close by the end of 2026, subject to customary closing conditions. The first leases are scheduled to begin in the first half of 2027, as the GPUs and supporting infrastructure start arriving at data centers selected by SpaceX. Apollo and Goldman Sachs are reported to have led the financing effort.
Earlier reporting from CNBC in October indicated that Apollo and a consortium of banks were in talks with SpaceX to arrange the financing, with the GPUs themselves expected to serve as collateral. The finalized lease agreement confirms that the private credit route won out over a conventional bond issue.
Nvidia Vera Rubin Chips at the Center of the Deal
Apollo said the financing will fund the delivery of Nvidia's next generation Vera Rubin chips for what it called SpaceX's terrestrial compute buildout. Vera Rubin is Nvidia's successor to the Blackwell platform and has not yet shipped at scale, which makes a $40 billion committed order book a meaningful signal of demand for hardware that is still ramping up production.
The new compute capacity is intended to power the Grok chatbot, enterprise and consumer AI services, and SpaceX's existing and future cloud customers. For Nvidia, the deal provides rare visibility into multi year demand from a single buyer with the financial strength to see the purchase through.
What SpaceX's Leadership Is Saying
SpaceX Chief Financial Officer Bret Johnsen said in a statement that the lease diversifies the company's funding sources and covers a notable share of the capital required to deliver on its growth plans. He added that SpaceX is seeing payback periods of less than one year on its AI infrastructure investments as it scales compute to support Grok and its cloud business.
Elon Musk, writing on X on Sunday, confirmed the financing and pointed to SpaceX's investment grade credit rating and the steady cash flow it earns from Starlink and rocket launches. He used the announcement to draw a sharp contrast with pure play AI compute providers, arguing that debt markets have largely shut out companies that lack significant revenue beyond AI data centers.
A Blockbuster Quarter for SpaceX's AI Business
The financing lands just a week after SpaceX reported a standout third quarter for its AI segment. That division, which includes Grok, the Cursor coding tool, and the company's cloud compute business, generated $7.6 billion of SpaceX's $14.6 billion in quarterly revenue. For the first time, AI brought in more money than Starlink and launch services combined.
Spending has climbed just as fast. Capital expenditures reached nearly $20 billion in the third quarter, up from $6.9 billion in the same period a year earlier. On the October 20 earnings call, Johnsen told investors that SpaceX expects to spend between $35 billion and $40 billion in the fourth quarter alone, with the vast majority going toward compute.
Why the Timing Matters
The agreement arrives after a difficult stretch in the debt markets for AI infrastructure companies. Investors have grown cautious about runaway capital spending paired with thin profits, and several so called neocloud operators that rent out GPU capacity have faced rising credit costs and canceled offerings in recent weeks.
SpaceX's ability to raise $40 billion with relative ease underscores a broader shift in how AI capital is flowing. Lenders are increasingly favoring operators with diversified, creditworthy revenue streams, while companies that depend entirely on GPU rental income are finding it harder to borrow. Musk predicted that the more than 200 neocloud companies in the market will face widespread failures and consolidation, with only a handful surviving.
What Comes Next
Whether the deal pays off will depend on execution. Key variables include how quickly Nvidia can deliver Vera Rubin systems, how fully SpaceX can utilize the hardware once it is installed, and whether AI and cloud revenue keeps growing fast enough to cover lease obligations as newer GPU generations arrive. With closing expected by year end and the first equipment landing in early 2027, the next six months will show how quickly SpaceX can turn committed capital into working compute.
Olivia Bennett is a technology writer at TechJox with over 8 years of experience creating detailed how-to guides, software tutorials, troubleshooting articles, and step-by-step walkthroughs. She specializes in Windows, Android, iOS, AI tools, software applications, browser settings, and everyday tech solutions. Her expertise includes simplifying technical processes, resolving common software issues, and explaining digital tools through practical, beginner-friendly instructions. Olivia focuses on making technology easier to navigate, helping readers solve problems confidently and get the most out of their devices and applications.
Discussion (0)
You need a free TechJox account to join the discussion and report scams you spot.