On the afternoon of July 23, Tencent announced the establishment of a basic model department, merging the big language model department and the multi-modal model department, and merging all related businesses and teams into the new department. Tencent's chief AI scientist Yao Shunyu is also the head of the basic model department and reports to Lu Shan, president of the Technology Engineering Business Group. Furthermore, Tencent did not announce the whereabouts of Bo Liefeng, the former head of the mixed-element multi-modal model department, and recently there was news that Hu Han, head of mixed element multi-modal understanding, has left his job. Former OpenAI researcher Tian Yonglong will take over Hu Han's job, but Tencent did not respond to this. After merging the big language model and multi-modal model departments, Yao Shunyu's responsibilities were expanded. Tencent said that this change will further improve the efficiency of model development and collaboration, and help explore the upper limit of intelligence for full-modal models.

On the afternoon of July 23, Tencent announced the establishment of a basic model department, merging the big language model department and the multi-modal model department, and merging all related businesses and teams into the new department. Tencent's chief AI scientist Yao Shunyu is also the head of the basic model department and reports to Lu Shan, president of the Technology Engineering Business Group. Furthermore, Tencent did not announce the whereabouts of Bo Liefeng, the former head of the mixed-element multi-modal model department, and recently there was news that Hu Han, head of mixed element multi-modal understanding, has left his job. Former OpenAI researcher Tian Yonglong will take over Hu Han's job, but Tencent did not respond to this. After merging the big language model and multi-modal model departments, Yao Shunyu's responsibilities were expanded. Tencent said that this change will further improve the efficiency of model development and collaboration, and help explore the upper limit of intelligence for full-modal models.