On March 20, the large model blind evaluation leaderboard LMArena updated its latest rankings. Alibaba's newest flagship preview model, Qwen3.5-Max-Preview, debuted on the list with a score of 1,464 points. The model outperformed leading overseas models such as GPT-4, Claude-4, and Grok-4, as well as all domestic models including Doubao 2.0, GLM-5, and Kimi 2.5.
In a separate LMArena ranking of global large model companies based on model capabilities, five Chinese companies made it into the top ten. Alibaba ranked among the top five globally and first in China. ByteDance, Zhipu AI, Moonshot AI, and Baidu also broke into the global top ten.
Comments