According to the "Kechuangban Daily" report on July 27, scholars from the University of California, Shanghai Jiaotong University, South China University of Technology and other universities recently jointly published a paper and launched a large model product called 3D-LLMs. The paper shows that 3D-LLMs can accept 3D point clouds and their features as input and perform a variety of 3D related tasks, including description, dense description, question answering, task decomposition, 3D positioning, 3D assisted dialogue, navigation, etc. Experiments show that the 3D-LLMs model outperforms the 2D VLMs model.