Paper Co-Authored by Tang Jie Reveals Zhipu's Internal Model GLM-Omni; Visual Generation May Be Next Variable for Next-Gen GLM

A research paper co-authored by Tang Jie has disclosed the existence of GLM-Omni, an internal multimodal model developed by Zhipu AI. The technical details suggest that the company is integrating advanced visual generation capabilities into its future model iterations. This development indicates a shift toward more comprehensive multimodal processing for the next generation of the GLM series.