Starcloud raises $170 million Series A to build data centers in the region | TechCrunch

Starcloud’s latest round of funding takes the space computing company to $1.1 billion, making it one of the fastest startups to reach unicorn status after receiving a Y Combinator grant.

The company’s Series A, which closed 17 months after its demo day offering, was led by Benchmark and EQT Ventures. It is another sign of interest in moving data centers around when resources and political obstacles slow down their development on Earth, but the business model is based on unproven technology and the use of large capital.

Starcloud has now raised a total of $200 million, and launched its first satellite with an Nvidia H100 GPU in November 2025. The company will launch a more powerful version, Starcloud 2, later this year with more GPUs, including an Nvidia Blackwell chip and an AWS server blade, as well as a bitcoin mining computer.

The company will begin developing a data center spacecraft designed to launch from Starship, a reusable heavy-lift rocket built by Elon Musk’s SpaceX. Starcloud 3, as the spacecraft has been named, will be a 200-kilowatt, three-ton spacecraft the size of a “pez dispenser” system SpaceX designed to launch its Starlink satellites from Starship.

CEO and founder Philip Johnston said that he expects it to be the first low-cost orbital data center and data centers in the world, with costs in the order of $ .05 per kw / hour of energy – if starting a business costs about $ 500 per kilogram.

The problem is that the Starship hasn’t flown yet; Johnston says he expects commercial access to open in 2028 and 2029. It’s a reality facing all major data center projects in space: powerful space computers will be expensive until a new generation of rockets begins to launch with high performance, something that may not happen until the 2030s.

“If it ends up being delayed, we will continue to produce smaller versions of the Falcon 9,” Johnston said. “We won’t be cost-competitive until Starships fly regularly.”

“There are two types of business,” Johnston explains: One is selling space power to other spacecraft; the company’s first satellite, for example, analyzes data collected by the Capella Space Radar spacecraft. Then, in the future when start-up costs come down, the most powerful distributed data centers can pull work from their global counterparts.

This finds out how new this industry is. When Nvidia CEO Jensen Huang unveiled the Vera Rubin Space-1 chip modules at his company’s annual GPU Technology Conference last week, he didn’t realize that none had been produced or shared with the company’s development partners.

In fact, the number of advanced GPUs in orbit is numbered, with Nvidia estimated to have sold 4 million hyperscalers worldwide by 2025.

Or consider that SpaceX’s Starlink network, which is the largest network of satellites in orbit with 10,000 spacecraft, produces about 200 objects. megawatts capacity, while data centers have more than 25 gigawatts power plants are currently being built in the U.S., according to Cushman and Wakefield.

Johnston argues that his company is ahead of the competition, with the world’s first GPU installed in orbit. It was used to train an AI model in orbit, the first, to follow the Starcloud, and to run the Gemini model. Beyond the process, Johnston says Starcloud now has key insights into what it takes to move a powerful chip into space.

“The H100 is probably not the best space chip, to be honest, but the reason we did it was we wanted to prove that we could run state-of-the-art chips in space,” he told TechCrunch. That hard-won knowledge — one GPU, the Nvidia A6000, failed during launch — will influence future designs.

There is a list of technical issues that need to be addressed, including efficient power generation and efficient chip cooling. Starcloud-2 will have the largest radiator that can be carried by an autonomous satellite; he expects at least two more types of spacecraft to go into orbit, Johnston said.

Then there is the challenge of coordination. The largest datacenter workloads, often for training, require hundreds or thousands of GPUs to work in tandem. Doing so in space would require an incredibly large spacecraft, or powerful and reliable laser communications between the spacecraft in flight. Many companies dealing with this technology expect those workloads to come long after simple learning tasks have come around.

In addition to Starcloud, Aetherflux, Google’s Project Suncatcher, and Aethero – which announced Nvidia’s first Jetson GPU for space in 2025 – are all companies that are developing a space data center.

The elephant in the room is SpaceX itself, which has asked the US government for permission to build and deploy a million satellites for distributed computing in space.

Dealing with SpaceX is a daunting task for any entrepreneur, but Johnston sees room for cooperation.

“They’re building a slightly different case than we are,” he told TechCrunch. “They’re mainly planning to serve a lot of Grok and Tesla. Maybe at some point they’ll offer third-party cloud services, but what I don’t think they’re likely to do is what we’re doing. [as] an energy and infrastructure player.”