China has unveiled the world’s first neuromorphic supercomputer capable of simulating more than 2 billion brain-like neurons, positioning itself at the forefront of energy-efficient artificial intelligence and brain-inspired computing. The new system, called Darwin Monkey (also known as Wukong), was developed through collaboration between Zhejiang University’s State Key Laboratory of Brain-Machine Intelligence and Zhejiang Lab.
Unlike traditional supercomputers, Darwin Monkey uses 960 specialized Darwin 3 neuromorphic chips, each supporting over 2.35 million spiking neurons, to reach a combined capacity of more than 2 billion artificial neurons and over 100 billion synapses. This immense computational scale mimics the function of the human brain by transmitting information through electrical spikes, closely resembling the way biological neurons operate. Despite its scale, the supercomputer operates with impressive efficiency, drawing only around 2,000 watts—far less than typical high-performance computing systems.
Central to Darwin Monkey’s design are Spiking Neural Networks (SNNs), which allow the system to process vast amounts of information quickly and efficiently. The supercomputer also incorporates the DeepSeek brain-like AI model, equipping it with advanced capabilities in reasoning, content creation, language comprehension, and mathematical problem solving.
Darwin Monkey’s versatility enables it to simulate the brains of organisms ranging from simple nematodes to more advanced species like macaques. This opens new avenues for neuroscience research, reducing reliance on live animal testing and offering researchers high-fidelity models of brain function.
One of the system’s most distinctive features is its complete reliance on domestically developed components—including chips, operating systems, and AI models. This self-sufficiency highlights China’s growing expertise and ambition in developing sovereign, end-to-end artificial intelligence infrastructure, a significant development amid ongoing global competition in advanced technologies.
Darwin Monkey sets a new benchmark in neuromorphic computing, surpassing Intel’s Hala Point neuromorphic platform, which previously led the field with its ability to simulate 1.15 billion neurons. With its double capacity, the Chinese innovation cements the nation’s leadership in energy-efficient, brain-like AI systems.
The debut of Darwin Monkey signals a shift in the direction of artificial intelligence research. While mainstream AI today relies on transformer-based architectures, neuromorphic systems inspired by the human brain offer a more sustainable and biologically plausible alternative. These new systems promise enhanced efficiency, adaptability, and the potential to reshape future approaches to both artificial intelligence and neuroscience.
Chinese experts emphasize that the achievements of Darwin Monkey are supported by thorough technical validation and have been officially confirmed by leading scientific institutions. The project reflects China’s determination to pioneer a new era in intelligent computing by developing technologies that mirror the remarkable efficiency and capability of the human brain.
