Alibaba officially launched its next-generation flagship AI model, Qwen4-Max, on November 16, 2026.
Qwen4-Max has 2.4 trillion parameters and uses a Mixture-of-Experts (MoE) architecture. It supports a 1 million-token context window and native visual understanding. Alibaba designed the model for complex tasks, including coding, agentic workflows, and professional applications.
The model is available through the API on the Qianwen AI Platform. Compared with its predecessor, API prices are significantly lower, with average per-task costs dropping by approximately 60%.
The launch follows a limited-time preview that began on November 11. The release positions Alibaba to compete more aggressively in the global generative AI and cloud computing markets.