China's AI compute race is shifting to supernodes, as cloud providers and model developers seek domestic infrastructure capable of handling surging large-model training and inference demand.
The article requires paid subscription.
Subscribe Now