Engineers at Zhejiang University in China have introduced a brain-inspired computer called Darwin Monkey, featuring over 2 billion artificial neurons.

Zhejiang University is a research institution based in Hangzhou, China.

According to the university, the system consists of 960 Darwin 3 neuromorphic chips, creating more than 100 billion synapses.

The neuron count is comparable to that of a macaque and is intended to advance brain-inspired AI research.

The Darwin Monkey has reportedly completed tasks such as content generation, logical reasoning, and mathematics, using a large AI model from Chinese company DeepSeek.

Developers say the system can simulate animal brains, including those of macaques, mice, and zebrafish, which may benefit neuroscience research.

Neuromorphic computing aims to mimic human brain function for more efficient processing, using spiking neural networks.

Intel previously announced a neuromorphic system with 1.15 billion neurons in 2023, deployed at Sandia National Laboratories in the US.

The Darwin 3 chip was developed by Zhejiang University in collaboration with Zhejiang Lab, which is funded by the Zhejiang provincial government, the university, and Alibaba Group, and each chip supports over 2.35 million artificial neurons.

🔗 Source: South China Morning Post

🧠 Food for thought

1️⃣ China’s neuromorphic breakthrough signals intensifying global competition in brain-inspired computing

The Darwin Monkey’s 2+ billion neurons represents a significant leap beyond Intel’s current industry-leading Hala Point system, which features 1.15 billion artificial neurons1.

This achievement demonstrates how quickly the neuromorphic computing field is advancing, with Chinese researchers now claiming the world’s largest brain-inspired system just months after Intel’s April 2024 announcement.

The rapid progression suggests that neuromorphic computing has moved beyond experimental research into a competitive technology race between major powers.

Intel’s Hala Point already demonstrated the potential of this approach, performing AI workloads 50 times faster than traditional systems while consuming 100 times less energy1.