Tongyi Lab has launched a major wave of AI updates, headlined by the open-source Qwen-Image-2512 model, which delivers unprecedented photorealism and text rendering for text-to-image generation. Alongside this, the lab released MAI-UI, a new foundation for GUI agents designed for complex mobile and desktop automation. The update also includes Qwen Code v0.6.0 with multi-provider support and advanced multimodal retrieval models.

These releases are complemented by MobileWorld, a rigorous new benchmark for testing mobile agent capabilities, which reveals current limitations in real-world deployment. This aggressive push into open-source tooling and foundational models positions Tongyi as a key player for developers building next-generation AI applications.