Alibaba has officially launched Qwen4-Max, its latest AI model, claiming it can outperform Claude and Kimi in coding tasks.
The model has 2.4 trillion parameters, but only uses 71 billion at a time to generate answers. It can also process up to 1 million tokens of information in one go and understand images.
Alibaba says Qwen4-Max performs better than Claude Fable 5.1 and Kimi K3.1 Pro in its coding tests. It also ranks 6th on Code Arena.
The performance claims come from Alibaba’s own tests