Alibaba Unveils New AI Chip and 20 GW Cloud Plan

Editorial illustration of an AI processor and engineers inside a large cloud data center.

Alibaba unveiled the Zhenwu V900 artificial intelligence processor at its Apsara Conference in Hangzhou on September 22, expanding a hardware strategy built around custom chips, cloud infrastructure and large AI models. Chief Executive Eddie Wu said the processor delivers three times the performance of its Zhenwu M890 predecessor. Alibaba described the V900 as China’s most powerful AI chip, a company claim that has not yet been established by independent benchmark testing. Reuters, September 22, 2026.

The new Zhenwu V900 AI chip carries 216 GB of memory and 1,200 GB per second of chip to chip bandwidth, according to specifications reported after the conference. Native support for FP8 and FP4 gives the processor low precision formats commonly used to accelerate AI workloads. Commercial release and mass production are planned for the first quarter of 2027. CRN Asia, September 22, 2026.

Alibaba said a single V900 cluster can scale to as many as 500,000 accelerator cards for model training and inference. The company is pairing its processor roadmap with networking, storage and server technology, giving Alibaba Cloud AI infrastructure more control over the computing stack rather than relying only on outside chip suppliers. Alibaba Cloud, September 22, 2026.

Cloud capacity is also scheduled for a major expansion. Alibaba set a goal for global data center capacity operated by Alibaba Cloud to exceed 20 gigawatts by 2032. The target represents planned capacity rather than computing power already deployed. Reaching that scale would require a large buildout of electrical, cooling, networking and server infrastructure over the next six years. TechNode Global, September 22, 2026.

The hardware plan is tied to a larger model roadmap. Alibaba said Qwen 4 is in training and future Qwen models could reach 5 trillion to 10 trillion parameters. Larger parameter counts do not automatically guarantee better model quality, but models at that scale can require enormous computing and memory resources during training and operation. Associated Press, September 22, 2026.

Alibaba’s expansion arrives as Chinese technology companies develop domestic alternatives to advanced processors affected by United States export controls. The company’s strategy links semiconductor design, cloud capacity and model development under one organization, creating a vertically integrated path from silicon to AI services. Investors reacted to the announcements by sending Alibaba shares higher after the conference presentation. Reuters, September 22, 2026.


BitcoinVersus.Tech Editor’s Note:
We volunteer daily to ensure the credibility of the information on this platform is Verifiably True. If you would like to support to help further secure the integrity of our research initiatives, please donate here: 3C9o19EH5HSiwEPyCTmEKzxhNCbo2X6TTb

https://x.com/1BitcoinVersus/status/1937006164555993338

BitcoinVersus.tech is not a financial advisor. This media platform reports on financial subjects purely for informational purposes.

Leave a comment