Alibaba's latest AI model ran autonomously for 35 hours to optimize code for its own custom chip

Alibaba’s Qwen team has unveiled its latest AI model, Qwen3.7-Max, which has demonstrated the ability to operate autonomously for an impressive 35 hours while optimizing code for its custom chip. This development is significant as it positions Alibaba at the forefront of AI capabilities, particularly in autonomous agent tasks, and highlights the growing competition in the AI landscape, especially against established players like Claude Opus and emerging Chinese rivals.

The Qwen3.7-Max model has shown competitive performance, matching Claude Opus 4.6 on key benchmarks and outperforming local competitors such as DeepSeek V4 Pro and Kimi K2.6. This achievement underscores Alibaba’s commitment to advancing AI technology and its practical applications. The model’s capabilities were further showcased through a demonstration involving a four-legged robot, illustrating its potential for real-world applications beyond mere code optimization. The implications for users are substantial, as this technology could lead to more efficient and autonomous systems across various industries, enhancing productivity and reducing the need for human intervention in complex tasks.

As Alibaba continues to refine its AI offerings, the market can expect increased pressure on competitors to innovate and improve their own models. The success of Qwen3.7-Max may prompt other tech giants to accelerate their AI development efforts, potentially reshaping the competitive landscape in the industry. Observers should keep an eye on how Alibaba leverages this technology in its broader product ecosystem and the responses from its rivals in the coming months.