Huawei’s Ascend-powered CloudMatrix computing platform tops biennial tech innovation list


Originally designed with 384 Ascend 910C neural processing units and 192 Kunpeng server central processing units, which are interconnected through a unified bus providing ultra-high bandwidth and low latency, the system can support up to tens of thousands of interconnected chips to train and run AI models, according to Huawei executives at the company’s sixth Innovation and IP Forum in Beijing on Tuesday.

The top invention integrated innovations in hardware, software and architecture, according to Sun Hongwei, Huawei’s chief computing software architect.

Sun said his team tackled plenty of challenges to maintain reliability, power supply and data storage for large computing systems.

A section of Huawei Technologies’ CloudMatrix 384 system, powered by its Ascend 910 AI processors, is displayed at the company’s exhibition space at the World Internet Conference in Wuzhen, eastern Zhejiang province, on November 8, 2025. Photo: Reuters
A section of Huawei Technologies’ CloudMatrix 384 system, powered by its Ascend 910 AI processors, is displayed at the company’s exhibition space at the World Internet Conference in Wuzhen, eastern Zhejiang province, on November 8, 2025. Photo: Reuters
  • Related Posts

    How Changan Automobile’s design-driven brand strategy shapes its global expansion plans

    Long seen as one of China’s traditional “big four” state-owned vehicle manufacturers, Changan Automobile, which has a history that dates back to 1862 but began to produce its own eponymous-branded…

    Continue reading
    Alibaba teases new Qwen previews, highest-ranking Chinese AI models on Arena

    The new model previews, the Qwen3.7-Max-Preview and the Qwen3.7-Plus-Preview, ranked 13th globally in text capabilities and 16th in vision capabilities, respectively, benchmark firm LM Arena said on Tuesday. The versions…

    Continue reading

    Leave a Reply

    Your email address will not be published. Required fields are marked *