Today Qwen3.7-Plus, the latest multimodal model in the Qwen 3.7 series, was
released and deployed on Qianwen Cloud and Alibaba Cloud Bairun; developers can
access it via API. Text and vision capabilities are materially improved — the
model ranks in the global top five and first in China on third-party benchmark
Vision Arena. Qwen3.7-Plus integrates visual perception with code generation,
tool invocation and GUI control, enabling closed-loop execution of complex tasks
within a single agent loop.
