Release of Fugaku-LLM – a large language model trained on supercomputer Fugaku


Researchers in Japan have developed Fugaku-LLM, a large language model with enhanced Japanese capabilities, using the RIKEN supercomputer Fugaku. The model has 13 billion parameters and outperforms other Japanese models, with potential applications in research and business, including AI for science and social simulation of virtual communities. [summary] [comments]


这是一个从 https://www.fujitsu.com/global/about/resources/news/press-releases/2024/0510-01.html 下的原始话题分离的讨论话题