China Launches Groundbreaking Space-Based AI Network with Three-Body Computing Constellation

On May 14, 2025, China launched the first 12 satellites of its ambitious Three-Body Computing Constellation from the Jiuquan Satellite Launch Center aboard a Long March 2D rocket. This initiative, led by ADA Space and Zhejiang Lab, aims to establish a space-based supercomputing network comprising 2,800 satellites.

The Three-Body Computing Constellation represents a significant leap in space-based computing and artificial intelligence integration. By processing data directly in orbit, this project reduces reliance on ground-based infrastructure, enhances real-time data analysis capabilities, and positions China at the forefront of space-based AI technology.

The constellation is a collaborative effort between ADA Space, a commercial aerospace company established in 2018 focusing on AI-driven satellite technology, and Zhejiang Lab, a research institute founded in 2017 by the Zhejiang provincial government, Zhejiang University, and Alibaba Group. The project's name draws inspiration from the "three-body problem" in physics and Liu Cixin’s science fiction trilogy, reflecting the complexity and ambition of managing a vast network of interconnected satellites.

Each of the 12 satellites launched is equipped with an 8-billion-parameter AI model capable of performing 744 tera operations per second (TOPS). Collectively, this initial batch delivers a combined computing capacity of 5 peta operations per second (POPS). The satellites utilize high-speed laser inter-satellite links, operating at 100 gigabits per second, to form a self-healing mesh network that enables ultrafast data transfer. One satellite also carries a cosmic X-ray polarimeter developed by Guangxi University and the National Astronomical Observatories of the Chinese Academy of Sciences, designed to detect and classify transient cosmic events such as gamma-ray bursts.

The primary goal of the Three-Body Computing Constellation is to perform real-time data processing in orbit, addressing efficiency issues in traditional satellite systems and accelerating the integration of artificial intelligence into space-based applications. By processing data directly in space, the constellation aims to reduce reliance on Earth-based data centers, overcoming bandwidth limitations and latency constraints associated with transmitting large volumes of data to the ground. This approach is expected to enhance applications such as astronomical observations, environmental monitoring, and cross-orbit communication.

The project also reflects China's growing ambitions to extend its computing power into space, an emerging frontier that could eventually complement terrestrial cloud infrastructure and support data-intensive applications like AI and remote sensing. The constellation's zero-carbon footprint, achieved through reliance on solar energy and the natural cooling environment of space, aligns with China's "New Infrastructures" campaign, emphasizing sustainable technological development.

Wang Jian, director of Zhejiang Lab and founder of Alibaba Cloud, highlighted the significance of the project, stating, "It's a good time to think about how we can put AI into space, not just in your laptop or cellphone. Space has, again, become the frontier for us to think about what we can do in the next 10, ... ."

While the United States and Europe have conducted tests on space-based computing, China's initiative represents the first operational deployment of a space-based AI network at this scale. This development underscores strategic competition with U.S. and European rivals, who have tested edge computing in space but lag in operational scale.

The successful launch of the initial satellites marks a significant milestone in China's pursuit of advanced space-based computing capabilities. As the Three-Body Computing Constellation continues to expand, it is poised to revolutionize data processing and analysis, offering new opportunities for scientific research, environmental monitoring, and global communication.