Alibaba Qianwen announced the official release of the Qwen 3.5 mid-sized model series, which includes Qwen3.5-Flash, Qwen3.5-35B-A3B, Qwen3.5-122B-A10B, and Qwen3.5-27B. According to the announcement, the new models offer enhanced intelligence and lower computational requirements. The performance of Qwen3.5-35B-A3B has surpassed that of Qwen3-235B-A22B-2507 and Qwen3-VL-235B-A22B—proving that architecture, data quality, and reinforcement learning are the keys to intelligence advancement, rather than simply stacking parameters. Qwen3.5-122B-A10B and 27B have further narrowed the gap between mid-sized models and state-of-the-art models, performing particularly well in complex agent tasks. Qwen3.5-Flash is a managed production version based on the 35B-A3B model, supporting a default context length of 1 million tokens and featuring a built-in official toolset.