This article first appeared on GuruFocus.
Alibaba Group (BABA, Financials), is deepening its artificial intelligence push with a much larger AI model, a new chip and an ambitious expansion of data-center capacity.
At Alibaba Cloud’s annual Apsara conference, CEO Eddie Wu said the company’s Qwen team plans to develop models with 5 trillion to 10 trillion parameters, compared with 2.4 trillion for its current Qwen 3.8 Max flagship.
Alibaba is also developing Qwen 4, with future Qwen 4.5 and Qwen 5 models expected to scale toward the 5 trillion-to-10 trillion range. The company said its models are making progress in improving themselves with less human involvement.
On the hardware side, Alibaba unveiled the Zhenwu V900, a new AI chip developed by its T-Head semiconductor unit. Wu said it delivers three times the performance of the previous M890 and can be connected in clusters of up to 500,000 chips.
The chip is expected to enter mass production and commercial release in the first quarter of 2027.
Alibaba Cloud also plans to push global data-center capacity beyond 20 gigawatts by 2032. Wu said AI demand is accelerating cloud revenue growth, although supply-chain constraints are limiting expansion.
Alibaba shares rose 5.1% Tuesday, showing how closely investors are watching the company’s progress in AI infrastructure and software.
The next catalyst is the commercial rollout of the Zhenwu V900 and Alibaba’s next-generation Qwen models.