Alibaba releases upgraded AI model Qwen 2.5-Max to compete with DeepSeek and others

Alibaba urgently launched an upgraded version of its AI model Qwen 2.5-Max during the Spring Festival to cope with rapidly changing industry competition. The company claims that the model "almost completely outperforms" ChatGPT, DeepSeek, and Meta's Llama in terms of performance. According to Alibaba's testing, Qwen 2.5-Max achieved a high score of 89.4 in the Arena Hard benchmark test and performed better than DeepSeek in the MMLU Pro university level problem-solving ability test, comparable to ChatGPT. Alibaba developers stated on GitHub that the new model is expected to surpass human intelligence and explore unknown territories. Former US President Trump stated that the strong performance of Chinese AI models should be a "wake-up call" for Silicon Valley, but he also believes that the rise of DeepSeek is an opportunity to reduce costs.