On Monday, Foxconn, the renowned Taiwanese electronics manufacturer, announced the launch of its inaugural large language model, aimed at enhancing its manufacturing processes and supply chain operations.
The model, named “FoxBrain,” was developed using 120 Nvidia H100 GPUs, with the training process concluding in approximately four weeks, according to a statement from the company.
Foxconn, which partners with Apple for iPhone assembly and produces artificial intelligence servers for Nvidia, has indicated that FoxBrain is based on the Llama 3.1 architecture from Meta.
This model marks Taiwan’s first large language model equipped with reasoning capabilities, specifically optimized for traditional Chinese and Taiwanese linguistic styles, Foxconn reported.
While there exists a minor performance disparity when compared to China’s DeepSeek’s distillation model, Foxconn contends that FoxBrain’s performance remains closely aligned with global standards.
Originally designed for internal use, the FoxBrain model encompasses various functionalities, including data analysis, decision support, document collaboration, mathematical computations, reasoning, problem-solving, and code generation.
Foxconn intends to partner with technology firms to broaden the applications of this model, share its open-source resources, and advance the implementation of AI within manufacturing, supply chain management, and intelligent decision-making.
Dedicating resources to this initiative, Nvidia has facilitated the process with its Taiwan-based supercomputer, “Taipei-1,” offering technical guidance throughout the training of the model. Taipei-1 is recognized as the largest supercomputer in Taiwan and is operated by Nvidia in the city of Kaohsiung.
Further details about the FoxBrain model will be disclosed during Nvidia’s GTC developer conference scheduled for mid-March.
© Thomson Reuters 2025
(This story has not been edited by NDTV staff and is auto-generated from a syndicated feed.)
