SpaceX has announced a partnership with NVIDIA to design a computing power deployment solution for satellites.

During an investor conference call on August 4, following SpaceX's first quarterly earnings report since its IPO, CEO Elon Musk revealed the company has decided to exclusively use NVIDIA technology for its AI computing infrastructure.

Musk stated that SpaceX considers NVIDIA's Vera Rubin architecture "the best AI computer currently available" and plans to begin launching Starmind satellites equipped with NVIDIA's Vera Rubin NVL72 rack systems next year, extending AI computing capabilities to low Earth orbit. This news boosted NVIDIA's after-hours stock price by more than 2%.

Exclusive Bet on NVIDIA, Starmind Satellite Launch Set for Next Year

Musk made it clear during the call that all future AI data center construction at SpaceX will exclusively use NVIDIA systems. He said: "We have established a close cooperation and partnership with NVIDIA at multiple levels, and we have chosen to use NVIDIA exclusively."

The two companies announced on August 4 that they will jointly develop the computing payload for the Starmind AI1 satellite, a spacecraft designed specifically for AI computing tasks in low Earth orbit, which will carry NVIDIA's Vera CPU and Rubin GPU.

Musk noted that the NVL72 system's design offers advantages over traditional standard rack configurations, so SpaceX plans to deploy the system both on the ground and in orbit simultaneously. SpaceX's orbital computing vision extends far beyond this, with the company envisioning a network of up to one million satellites interconnected via laser links, forming a distributed space-based AI supercomputer.

This partnership also marks SpaceX's official entry into NVIDIA's space computing ecosystem. When NVIDIA launched its space computing platform in March, the six launch partners announced did not include SpaceX.

Computing Expansion Accelerates, Targeting 20 Gigawatts

In terms of computing scale, Musk provided more aggressive expansion targets. He indicated that SpaceX expects its online computing capacity to exceed 2 gigawatts by the end of this year, and approach 10 gigawatts by the end of next year, rather than the previously anticipated 5 gigawatts. When Morgan Stanley analyst Adam Jonas asked about the pace of computing construction, Musk stated the company's actual targets have surpassed previously disclosed ranges.

He said: "We hope to reach a cumulative 20 gigawatts by the end of next year through a series of projects." Musk also noted that current resource allocation priorities will focus on power and cooling equipment, rather than GPUs themselves. While acknowledging some projects may not be fully completed on schedule, Musk estimated that approximately 15 gigawatts of online computing capacity could be achieved by the end of 2027.

Orbital data centers are considered a potential alternative to large ground-based facilities. Compared to terrestrial installations, space deployment requires less land and faces lower cooling pressure, but sending servers into orbit and establishing network connections presents equally significant engineering challenges.