FOURTH PARADIGM (06682) today launched its "ModelHub XC" platform for indigenous innovation, along with a dedicated community and value-added services for model adaptation. The launch event was witnessed by multiple leading chip manufacturers including Huawei Ascend, BiRen Technology, Tianshu Zhixin, Kunlun Core, Moore Threads, and Cambricon, as they collectively work to resolve compatibility issues between domestic AI models and domestic chips.

Within the domestic computing ecosystem, models and chips frequently face compatibility problems, resulting in lengthy deployment cycles and substantial redundant workload. FOURTH PARADIGM has introduced the EngineX AI engine system specifically adapted for indigenous computing power. Through fundamental algorithm architecture adaptation, it enables batch model support, shortens deployment cycles, and achieves "engine-driven, plug-and-play multi-model" functionality.

For enterprise-level requirements, FOURTH PARADIGM provides comprehensive value-added services backed by hundreds of engineers, ensuring model adaptation to designated domestic computing platforms. The company has currently adapted and certified over one hundred models, with expectations to reach tens of thousands within one year and achieve coverage of all common computing platforms within six months.

The launch of FOURTH PARADIGM's "ModelHub XC" platform breaks the binary separation between "computing power and models," potentially providing crucial support for the large-scale implementation of the AI industry.