Starcloud raises $250M for orbital data centers as launch supply tightens

Starcloud, a startup building satellites that run AI inference in orbit, has added a $250 million extension to its March $170 million Series A round, according to TechCrunch. The extension values the company at $2.3 billion.

The new capital will fund a larger manufacturing facility and advance Starcloud-3, the company's largest orbital data center spacecraft, which is designed to fly on SpaceX's forthcoming Starship rocket. Starcloud has already requested FCC permission to operate 88,000 spacecraft.

We can see what's coming. We're going to need to book an enormous amount of launch.

Launch capacity across the industry is tightening. SpaceX's workhorse Falcon 9 program is scheduled to end in 2028, and the company is shifting toward the larger but still unproven Starship. Competing rockets aren't ready to fill the gap yet: Blue Origin's New Glenn and ULA's Vulcan aren't flying regularly, and Rocket Lab's Neutron isn't on the pad. Johnston pointed to launch access as one of the company's biggest cost pressures.

One of the biggest costs is now on securing your launch capacity. Launch is pretty constrained right now because Falcon 9 program is scheduled to end in 2028.

In the near term, Starcloud is focused on launching its new generation of 8kW compute satellites, called Starcloud-2, on rideshare flights in 2027 to perform orbital inference for customers including US government agencies. The company is also weighing a dedicated Falcon 9 launch and contracts with other providers to support future missions. Johnston said he wants to lock in Starship access as soon as the vehicle is ready.

• Funding round: $250 million extension, $2.3 billion valuation
• Lead investor: Manhattan West Ventures, with Nvidia and Cisco participating
• Nvidia's contribution: about $25 million, per a person familiar with the deal
• Current hardware: the only known company running an Nvidia H100 GPU in orbit
• Facility: 100,000 square feet in Woodinville, Washington, 25 employees

Nvidia's participation stood out to Johnston as a signal of Starcloud's position in the young space-compute sector. Starcloud says it's the only company currently operating an Nvidia H100, a terrestrial-grade data-center GPU, in orbit, and the first to train a model using one in space. Most other space-based GPUs handle lighter edge processing rather than full data-center-class training workloads.

The reason they've chosen to do this investment now is because of all of this data that we got from Starcloud One. They, more than any other VC, did way more technical duty on this than anybody else.

Starcloud is sharing what it has learned in orbit with Nvidia as the chipmaker develops its first purpose-built space GPU, the Vera Rubin Space-1. The chip hasn't been built yet, and Starcloud hopes to fly it into orbit sometime in late 2028. Johnston said his engineers are tracking a handful of design tradeoffs: the relationship between a chip's running temperature and the size of the radiators needed to dispel that heat, where to place radiation shielding, and how to ruggedize the hardware to survive a rocket launch.

Starcloud's push mirrors a broader pattern of capital flooding into AI compute this year, a dynamic that also shows up in how Meta pays Microsoft hundreds of millions for Azure AI access even while building its own competing systems. Starcloud is currently 25 employees strong and building production lines at a 100,000-square-foot facility in Woodinville, Washington, near where SpaceX and Amazon build satellites for their own communications networks.

Nothing here should be taken as financial advice — just information to consider.