TMTPOST — Alibaba officially launched its Qwen3.7-Plus multimodal AI agent model on Tuesday, introducing a major upgrade to its proprietary foundational intelligence stack. Building on the text capabilities of its core Qwen3.7 architecture, the new iteration features an overhauled vision-language framework designed to unify visual perception with textual processing. According to the company's official announcement, the model retains full "agentic" capabilities, focusing heavily on executing multi-step workflows across coding, tool usage, and enterprise productivity tasks. The release marks a targeted push by the e-commerce and cloud giant to deploy cost-effective, full-modality interactive tools that seamlessly bridge graphical and command-line automation interfaces. The roll-out positions Alibaba to capture critical market share in the rapidly expanding enterprise automation landscape. By maintaining native support for developer toolchains while expanding visual comprehension, the company aims to embed its models deeply into complex backend software engineering pipelines.