July 21 — Alibaba released Qwen-Image-3.0, the third-generation foundation model in its Qianwen image-generation and editing series. The model supports up to 4.5k-token inputs and can generate multi-element knowledge diagrams combining formula notation, geometric figures and logical derivation steps, as well as complex UI mockups. It supports 12 languages and native rendering for more than 20 fonts.(金十数据)