Shaohua Wu, Xudong Zhao, Tong Yu, Rongguo Zhang, Chong Shen, Hongli Liu, Feng Li, Hong Zhu, Jiangang Luo, Liang Xu, Xuanwei Zhang.

September 28, 2021 marked the release of Inspur's Yuan 1.0, which upon release, became one of the world's largest language model with 245.7 billion parameters and 5 TB of datasets. The following technical paper, also published on arXiv, details how Inspur managed to create such a large model, and how it was able to eliminate processing restraints normally expected at this size.