SpaceX targets a 5x compute capacity increase by end of 2027, requiring procurement of over 2 million Nvidia Rubin GPUs—an unprecedented volume.

SpaceX wants to grow its compute capacity more than fivefold by the end of 2027, aiming to finish the year approaching ten gigawatts of power, according to Elon Musk. The company currently sits at 1.4 gigawatts. The entire buildout will run on Nvidia chips, specifically leveraging Nvidia's upcoming Vera Rubin platform, which Musk described as the best architecture available. Existing Colossus clusters currently run on Nvidia's H100, GB200, and GB300 systems.

SpaceX's IPO filings detail its current and planned Colossus buildout at roughly 100,000 H100 units, 110,000 GB200 processors, and 110,000 GB300 processors, plus at least 220,000 additional planned GB300 units. That totals approximately 540,000 GPUs. The new capacity targets would push the overall number well past one million Rubin GPUs, potentially exceeding two million if the entire expansion uses Vera Rubin processors. SpaceX has not specified what proportion of the buildout will actually be Rubin chips.

Musk's AI company xAI merged into SpaceX in February 2026 through a stock-funded deal that valued SpaceX at $1 trillion and xAI at $250 billion, for a combined valuation of $1.25 trillion. Long-term, Musk envisions operating data centers in orbit.

SpaceX's AI segment, which includes xAI and X, posted $2.56 billion in revenue during Q2 2026 alongside an operating loss of $1.26 billion. In Q1, the segment had lost $2.47 billion on $818 million in revenue. For the first half of 2026, operating losses in the segment totaled $3.73 billion. Most of the revenue growth did not come from Grok but from new cloud contracts leasing Colossus compute capacity. A single unnamed AI customer accounted for roughly $1.52 billion of that revenue—likely Anthropic.

Competitors diversify their compute infrastructure across multiple providers. Anthropic uses AWS Trainium, Nvidia GPUs, and Google TPUs to train and run Claude, with publicly announced plans for up to five gigawatts of additional AWS infrastructure using Trainium and Graviton systems, plus approximately 3.5 gigawatts of extra TPU capacity. OpenAI has announced six gigawatts of AMD capacity, at least ten gigawatts of Nvidia capacity, and around two gigawatts of AWS Trainium capacity.