Qwen: Qwen Plus 0728 (retires Oct 9)

Built upon the Qwen3 foundation, this model is designed to serve as a versatile, balanced solution for users who require a middle ground between high-capability and high-speed processing. Its architecture is specifically engineered to handle extensive text-based tasks, providing a robust framework for applications that demand both depth and efficiency. By focusing on a stable, snapshot-based design, it ensures consistent behavior for production environments, making it a reliable choice for developers and teams looking to integrate sophisticated text generation and structured output capabilities into their workflows.

The model incorporates hybrid reasoning capabilities, allowing users to toggle between standard and thinking modes to better suit the complexity of their specific queries. This flexibility, combined with its strong performance in math, coding, and logical reasoning, positions it as a significant advancement over earlier iterations. With its ability to support diverse multilingual requirements and industry-leading tool calling, the model is well-suited for complex agentic workflows and creative writing projects, providing a scalable and cost-effective tool for modern AI-driven applications.