Foxconn developed a language model to improve production and supply chain management

Taiwanese company Foxconn has developed its own large language model (LLM) called FoxBrain and plans to use it to improve production and supply chain management, Reuters reports.

The model is based on Meta’s Llama 3.1 architecture and was trained using 120 NVIDIA H100 GPUs. It is the first large-scale reasoning language model optimized for traditional Chinese and Taiwanese styles.

Foxconn said that while there is a slight difference in performance compared to the Chinese DeepSeek model created through distillation, its overall efficiency is very close to global standards.

Foxconn plans to collaborate with other technology companies to expand the application of FoxBrain , open source it, and advance AI in manufacturing, supply chain management, and intelligent decision-making.

Foxconn will announce additional details about the model during the NVIDIA GTC developer conference in mid-March.