Foxconn has recently unveiled a large language model designed to enhance its manufacturing and supply chain management processes. Named "FoxBrain," the model boasts capabilities in logical reasoning, data analysis, mathematical computation, and code generation.
The development of FoxBrain took four weeks, utilizing 120 Nvidia H100 GPUs. In a significant move, Foxconn plans to release the model as open-source, fostering collaboration with industry partners. According to the company, FoxBrain is built upon Meta's Llama 3.1 model and has achieved performance levels close to the global leading standard.
This step highlights Foxconn's deeper exploration into artificial intelligence, aiming to boost production efficiency and advance intelligent supply chain management through technological innovation.
