Alibaba Group revealed China’s strongest AI chip to date alongside plans to develop an unprecedented 10-trillion-parameter AI model, reinforcing its push towards artificial superintelligence.

  • Alibaba introduces Zhenwu V900, China’s most powerful AI chip.
  • Plans to train Qwen 5 model with up to 10 trillion parameters.
  • Targets 20 gigawatts cloud capacity expansion by 2032.

What happened

Alibaba Group showcased its latest advances in artificial intelligence at the annual Apsara Conference in Hangzhou. The company debuted the Zhenwu V900 AI processor, billed as China’s most powerful AI chip, delivering triple the performance of its predecessor. Alibaba also revealed its roadmap for the Qwen large language model series, with plans to scale up from the current 2.4 trillion parameter flagship to a future AI model featuring between 5 trillion and 10 trillion parameters.

The event highlighted Alibaba’s commitment to end-to-end AI development, incorporating cutting-edge AI models, chips, and cloud computing. Leadership underscored ongoing investments in AI cloud infrastructure, aiming to grow Alibaba Cloud’s global data-center capacity beyond 20 gigawatts by 2032, supporting machine intelligence at scale.

Why it matters

Alibaba’s unveiling marks a significant milestone in China’s race for AI supremacy, positioning the company at the forefront of both AI hardware and model development. The Zhenwu V900 chip’s advancement reflects China’s determination to reduce reliance on foreign semiconductors amid ongoing geopolitical restrictions, while pushing domestic innovation in AI processing capabilities.

Scaled-up language models with up to 10 trillion parameters represent important progress toward artificial superintelligence that could outperform human cognitive tasks. Alibaba’s integrated AI approach—spanning model architecture, chip design, and cloud deployment—aims to turn AI from experimental technology into widespread commercial value, ultimately changing how intelligence is commodified and applied.

What to watch next

Monitor Alibaba’s launch timeline for Zhenwu V900 mass production planned for early 2027 and the planned release of the next-generation Zhenwu J900 chip in 2028, alongside new custom CPUs optimized for AI. These hardware releases will be critical benchmarks for Alibaba’s competitiveness in China’s evolving AI semiconductor market, particularly against rivals like Huawei.

Watch for further announcements from Alibaba’s Qwen project on model capabilities and deployment, especially how its massive parameter models enhance applications in multimodal AI and complex task execution across devices. Additionally, verify how Alibaba’s collaborations with partners, including Apple, influence AI mobile ecosystems in China.

Source assisted: This briefing began from a discovered source item from SCMP China Tech. Open the original source.
How SignalDesk reports: feeds and outside sources are used for discovery. Public briefings are edited to add context, buyer relevance and attribution before they are published. Read the standards

Related briefings