Alibaba says its latest AI model, Qwen3.8-Max, outperforms startup Moonshot AI’s Kimi K3 on key benchmarks. Kimi K3, launched on July 16, has drawn global attention since DeepSeek’s R1 and is currently the top model to be toppled. Alibaba’s new flagship is tuned for autonomous coding, complex research, and other “long horizon” tasks. Like Kimi K3, it is a “mixture of experts” model that can interpret text, images, and video. The company wrote in a blog post that Qwen3.8-Max “self-evolves through feedback loops.” The new Qwen has 2.4 trillion parameters, compared with Kimi K3’s 2.8 trillion, and uses 95 billion parameters at a time versus Kimi K3’s 104 billion to control costs and response delays. Alibaba’s preceding flagship, Qwen 3.7, launched just two months before its successor, while Kimi K2 launched a full year before K3. Data scientist Mehul Gupta of DBS Bank said, “The speed at which the Qwen team ships models is starting to look ridiculous.”