How China’s new ‘Darwin Monkey’ could shake up future of AI in world first
First such supercomputer with over 2 billion artificial neurons mimics macaque brain, is expected to help advance human brain-inspired AI
Chinese engineers have unveiled the world’s first brain-like computer made up of more than 2 billion artificial neurons.
The Darwin Monkey is the latest generation of brain-inspired computers produced by Zhejiang University researchers.
“This is the world’s first brain-like computer based on a dedicated neuromorphic chip with more than 2 billion neurons,” the university said on its social media account on Saturday.
The computing system, made up of 960 Darwin 3 brain-inspired computing chips creating over 100 billion synapses, is “a step closer to achieving more advanced brain-like intelligence”, it said in the post.
The Darwin Monkey has been successfully deployed to complete tasks like content generation, logical reasoning and mathematics, using the groundbreaking Chinese AI company DeepSeek’s brain-like large model.
The neuron and synaptic resources of the brain-inspired computer could also be used to simulate the brains of various animals, such as macaques, mice and zebrafish, which the university noted could advance brain science research.
Neuromorphic computing, also known as brain-inspired computing, draws inspiration from the brain’s neural networks and processing capabilities through the use of artificial neurons and synapses.
Brain-inspired computing can allow for more efficient information processing by allowing systems to mimic cognitive functions like decision-making, learning and memory.
This could enable faster and more adaptable problem-solving as well as more advanced AI systems.
Spiking neural networks go one step further than traditional artificial neural networks, by using discrete spikes or signals rather than continuous values to represent and transmit data, thereby more closely modelling the function of biological neurons.
In April last year, Intel announced that it had built the first 1.15 billion-neuron neuromorphic computing system, Hala Point, saying it would build a path towards “more efficient and scalable AI”.
The system was initially deployed at the US Department of Energy-funded Sandia National Laboratories.
The Darwin 3 chip, which the Darwin Monkey system relies on, was developed in 2023 by Zhejiang University in collaboration with the research institute Zhejiang Lab. The lab is jointly funded by the university, the Zhejiang provincial government and Alibaba Group. The South China Morning Post is owned by Alibaba.
Each Darwin 3 chip supports more than 2.35 million spiking or pulsed neurons and hundreds of millions of synapses, and comes with “specialised brain-inspired computing instruction sets and neuromorphic online learning mechanisms”, the university said.
Under typical operating conditions, the system only consumes around 2,000 watts of power.
“[Darwin Monkey’s] large-scale, high parallelism and low-power features will provide a new computing paradigm for existing computing scenarios,” Pan Gang, director of the State Key Laboratory of Brain-Computer Intelligence at Zhejiang University, told ministry newspaper Science and Technology Daily on Saturday.
The Darwin Monkey is the outcome of breakthroughs in a number of technologies, including improving the interconnection and integration of the neural system and developing a new generation of brain-inspired operating systems, according to the university.
WAIC Shanghai: China reveals new great leap forward with 1,509 AI models
That number accounted for more than 40 per cent of the 3,755 total AI models known worldwide, according to data from the Shanghai conference
There are 3,755 total AI models known worldwide, according to a Xinhua report on Monday, citing WAIC data.
“We see AI enabling scalable and cost-efficient production of multimodal content across text, images, audio and video,” said UBS Securities analyst Wei Xiong, who pointed out that Chinese models were “showing early success in AI video generation”.
The number of AI models reflects the country’s big strides in transforming into an AI powerhouse through various private and publicly backed open-source development initiatives, narrowing the gap with the US.
The open-source approach gives the public access to a programme’s source code, allowing third-party software developers to modify or share its design, fix broken links or scale up its capabilities.
WAIC also showed a rising trend among domestic AI infrastructure suppliers to join forces in building commercial computing resources amid the strong demand for advanced computing power in China’s AI sector.
“The primary reasons are the market’s lack of confidence in new domestic products and the absence in China of a company comparable to Nvidia that can offer comprehensive, full-stack solutions,” Chen told the Post in an interview on Sunday.
“What China needs most now is unity – and that’s what we are good at.”
While the WAIC organiser has yet to disclose the number of visitors at this year’s event, this figure was expected to be higher than the 2024 event’s 300,000 visitors, which was a record high since the conference started in 2018.
