Alibaba has announced the Zhenwu V900, an AI training and inference accelerator that it describes as the most powerful AI chip in China. The chip carries 216 GB of GPU memory and 1,200 GB/s of inter-chip bandwidth. Alibaba says the V900 is due for mass production and commercial release in the first quarter of 2027, moving ahead of the third quarter of 2027 timeline in the company's May roadmap.
The V900 can scale out to a supernode cluster of up to 500,000 cards, according to Alibaba. Its predecessor, the M890, had 144GB of memory and 800GB/s of inter-chip bandwidth, and worked in a supernode of 128 chips. Alibaba's chip arm T-Head says more than 1,000 V900 chips can operate as a single system. Alibaba did not release FLOPS, a process node, foundry, power figures or memory bandwidth claims for the V900.
As of May, T-Head had shipped more than 560,000 Zhenwu chips to over 400 external customers, a count Alibaba now puts above 650. The company expects a Zhenwu refresh annually. Huawei's Ascend 960PR, due in the third quarter of 2027, is slated for 192GB of memory capacity, 2.4TB/s of memory bandwidth and a 2.2TB/s scale-up interconnect. Alibaba's performance claims currently rest on no absolute compute figure.
Alibaba Cloud's stated goal is a data center capacity of more than 20 gigawatts by 2032. Wu said global shortages in the AI supply chain are currently limiting how quickly the company can scale its compute infrastructure. Alibaba also anticipates significant growth in its own annual AI chip shipments, with domestic demand remaining high as Huawei had to shelve its global Ascend rollout.
On models, Alibaba said Qwen 4 is currently in training, with the Qwen 4.5 and 5 families to scale up further on the roadmap. The company claims Qwen3.8-Max ran 33 iterative cycles over a month of automated runs to produce an Artificial Analysis score gain from 40 to 45. A separate chip-design experiment with more than 10,000 EDA tool calls over more than 60 hours of test time showed an area cut of 42% without performance loss.
The announcements come ahead of Chinese President Xi Jinping's summit with President Donald Trump in Washington on Thursday, where AI remains a top agenda subject following U.S.-proposed AI mechanisms. The AI talks in New York did not have export controls for chips or tools on the agenda. The V900 is months from commercial release, and the 5-to-10-trillion-parameter model Alibaba promises remains untrained, with capacity targets six years out.
More AI news from TechManNews.
