Alibaba's Qwen3.7-Plus combines visual AI with autonomous agent capabilities for coding and app navigation

Summary

Alibaba's Qwen team has released Qwen3.7-Plus, a proprietary multimodal AI model that combines visual perception with agent capabilities like coding, tool use, and GUI navigation. Built on the text-only Qwen3.7, it functions as a "multimodal interactive hybrid agent" capable of recognizing real-world scenes, reading screens, operating interfaces, writing code from visual templates, and navigating mobile apps. In a demo, an agent built on the model autonomously developed a vocabulary learning app over eleven hours, producing over 10,000 lines of code across 1,000 agent calls. The model leads on-screen understanding in Qwen's benchmarks but shows mixed overall performance. It is priced well below Western frontier models and does not have open weights.

Source

bskyAlibaba's Qwen3.7-Plus combines visual AI with autonomous agent capabilities for coding and app navigationthe-decoder.com