The large model failed to solve elementary school math Olympiad problems. All nine AI large models performed poorly, revealing the limitations of large models. iFLYTEK and Baidu's models performed relatively well, while MiniMax ranked at the bottom in the test.