Elon Musk says ASML and other key semiconductor equipment makers have sped up deliveries for Terafab, the chip-manufacturing megaproject shared by Tesla and SpaceX. He made the remarks on October 20 during a live broadcast of SpaceX's third-quarter 2026 earnings call on X.
Musk said the main Terafab facility, in Gibbons Creek, Texas, is aiming for volume chip production in the third or fourth quarter of 2028. He described that schedule as highly aspirational, with "big error bars." "These are very aggressive dates, and we're very appreciative of ASML and all of our other suppliers in accelerating equipment deliveries in order to achieve this outcome," said Musk, who is CEO of both companies behind the project. SpaceX expects to make its first experimental test chip at its research fabrication site in Austin, Texas, around the second quarter of 2027.
He also laid out near-term targets for his AI businesses. Musk expects 2.5 to 3 gigawatts of computing capacity to be active by the end of 2026 and is highly confident of adding another 10 gigawatts in 2027. He said he would be surprised if the total did not reach 13 gigawatts by the end of 2027. By comparison, the restructured 10-gigawatt Project Stargate and Meta's 5-gigawatt Hyperion were both expected to reach full capacity in the 2030s.
To get there, Musk said SpaceX and xAI are aggressively buying Nvidia's next-generation hardware. He announced plans to bring a 440,000-GPU Nvidia "Vera Rubin" training cluster online by the end of the second quarter of 2027.
Musk also claimed that, once fully scaled, Terafab's chip output would be valuable enough to rank among the world's 20 largest national economies, all from a single building. The 20th-largest, Switzerland, has a GDP of roughly $1.14 trillion. Terafab is planned as a multibillion-dollar, vertically integrated factory aiming to produce 1 terawatt of AI computing power per year, with logic, memory and advanced packaging under one roof. It will use extreme ultraviolet lithography built around Intel's 14A manufacturing node. Intel joined the venture in April 2026 to help co-develop the fabrication technology.
ASML may be the project's most important partner, as the world's only commercial supplier of the roughly $400 million EUV systems needed for leading-edge, high-volume production. Unofficial sources have said Terafab's first EUV tool was slated for delivery in mid-2027. It is unclear whether that is the accelerated schedule Musk described. Other suppliers include Applied Materials, Lam Research, KLA and Tokyo Electron, which would provide front-end processing, metrology and process-control equipment. Tesla and SpaceX are also in talks with about 10 South Korean equipment makers, among them Hanmi Semiconductor and HPSP, as well as firms supplying specialized piping components and process valves.
Musk's comments come as capacity across the semiconductor supply chain remains tight. Terafab is meant to reduce the risk of future shortages of cutting-edge AI processors from outside foundries such as TSMC. Its chips would serve xAI's clusters, self-driving vehicles, Optimus humanoid robots and SpaceX's orbital AI data center, as well as the wider AI industry.
Many experts have questioned whether the 1-terawatt goal is realistic. Gartner estimates global AI-optimized data center capacity at about 44 gigawatts as of late 2026, which is 4.4% of Terafab's planned annual output. Epoch AI, meanwhile, says global AI computing capacity has grown by roughly 3.3 to 3.4 times per year since 2022.
