On November 16, Alibaba released the new generation of foundational large model, Qwen4-Max, featuring a total of 2.4 trillion parameters, native visual capabilities, and support for a context window of one million tokens. Built on an innovative model architecture and training optimizations, Qwen4-Max achieves higher performance with lower token consumption, continuing to evolve in key capabilities such as agents and programming: In the authoritative global front-end coding blind test CodeArena, Qwen4 ranked sixth; in third-party evaluations of agent capabilities and professional tasks, such as Agents’ Last Exam (ALE) and Vals Index, Qwen4 remained in the leading tier. Qwen4 is deeply optimized for real-world tasks across all domains, enabling better utilization of professional tools to handle long-cycle tasks and promoting the in-depth implementation of AI in specialized fields such as automotive manufacturing, game development, financial investment research, and scientific research.
Alibaba Officially Launches Qwen4: Enhanced Performance, Better Cost-Effectiveness, and 60% Reduction in Average Cost per Task
Recommended Readings
LongbridgeAI
