On August 5, according to Tencent Science and Technology, the Chinese multi-modal large model SuperCLUE-V benchmark list was released in August, and Tencent  Hunyuan large model ranked first in the domestic large model (71.95 points).Tencent claims that the model accurately identifies image elements and generates natural language descriptions, with comprehensive understanding and insight into details. The evaluation covered 12 large models of multi-modal understanding with high representation at home and abroad, and Tencent  Hunyuan large model obtained 71.95 points in the basic ability and application ability of multi-modal.

來源：DIYuan