On April 3, Code Arena—a global blind test ranking platform under LMArena that focuses on assessing AI programming capabilities—released its latest rankings. Alibaba’s next-generation large language model, Qwen 3.6-Plus, achieved a remarkable second-place finish globally, outperforming international tech giants such as OpenAI and Google. This marks Qwen 3.6-Plus as the highest-ranked Chinese large model to date. Notably, Qwen 3.6-Plus excelled in the React-specific ranking, showcasing its exceptional proficiency in code generation and engineering.