SuperCLUE October list released: GPT4 continues to dominate the list, vivo ranks first in China
According to news on the morning of October 26, SuperCLUE recently released the October list of Chinese large models. GPT4 ranked first in the overall list. Vivo’s self-developed large model ranked fourth in the overall list with a total score of 70.74, ranking first among domestic large models. First, followed by Moonshot, Wenxinyiyan 4.0 and SenseChat 3.0.
SuperCLUE is a multi-level comprehensive evaluation benchmark for Chinese general large models in the industry. It can be seen from the list that the competition among domestic large model products is still fierce, with vivo's self-developed large models leading the three lists (overall ranking, multi-round open question OPEN ranking, top ten basic capabilities list), Wen Xin Yi Yan 4.0 and SenseChat 3.0 have ranked among the top three in the OPT rankings for the three major ability objective questions. Each major model has shown relative competitive advantages in specific fields.
According to previous news, the 2023 vivo Developer Conference will officially open on November 1. During the conference, vivo will release 5 self-developed large model matrices, including three different parameter levels of billions, tens of billions, and hundreds of billions.