Recently, a tech enthusiast tested the programming performance of multiple AI models using LeetCode problem 3562 (stock trading discount problem). The results showed that Xiaomi MiMo-V2-Flash and Yuanbao model passed the test (AC) on the first attempt, demonstrating excellent performance; while DeepSeek, GLM4.6, and Doubao models all experienced timeouts, and the code generated by the Kimi model was completely unable to run. This comparative test highlights the efficiency and stability of Xiaomi’s AI model in solving complex programming tasks, providing valuable performance reference data for the AI industry and helping developers and tech enthusiasts choose the right AI tools.
