MedBench, a leading domestic evaluation platform for medical large models, has recently released its updated rankings. The medical large model from Ant AI Health Butler team has maintained its double championship status, scoring impressively high marks of 97.5 points in the evaluation ranking and 98.2 points in the self-test ranking. Building on this success, the Ant Health team has introduced a next-generation of medical reasoning models leveraging reinforcement learning. This advanced AI engine equips the large model with sophisticated medical thinking and reasoning abilities, enabling it to support multimodal interactions including images, text, audio, and video.
