Changes

To continuously enhance model performance and service experience, while optimizing resource management, Model Studio will decommission the snapshot models listed in the table below on 2026-05-09.

The Qwen3.5 native vision-language Flash models are built on a hybrid architecture that integrates a linear attention mechanism with a sparse mixture-of-experts model, achieving higher inference efficiency. Compared to the 3 series, these models deliver a leap forward in performance for both pure text and multimodal tasks, offering fast response times while balancing inference speed and overall performance.

Impact of the Change

After decommissioning, model inference will no longer be supported. Applications and services currently invoking these models will fail to return results properly. If you are using any of the above models, please update your applications before 2026-05-09 to avoid disruption to your production services.

Click here to view the Model Studio model list and pricing information.