Unverified50% confidenceReleaseExact time
Alibaba has released a GUI Agent model covering three major platforms—mobile, desktop, and web—capable of determining click and input operations by recognizing on-screen content.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/1/2026
First Seen
Valid until: 11/30/2026
Sources
Related Entities
Related Claims
Unverified计算机使用（GUI Agent）能力的底层原理是让模型通过截图理解屏幕状态并输出鼠标点击、键盘输入等操作指令，形成感知-决策-执行闭环72% similarUnverifiedAgent Device allows AI to perform actions including tapping, typing, and swiping on mobile devices.71% similarUnverifiedAgent Browser is a skill developed by the Vercel team that allows an agent to run a Chrome browser in the background to independently inspect the interface.71% similarUnverifiedAgentScope is a multi-agent framework developed and open-sourced by Alibaba's Tongyi Lab, used for building multi-agent applications in complex scenarios.71% similarUnverifiedTencent has launched an Agent that can autonomously help users operate various apps on their phones after learning their usage patterns.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/837555API
curl https://kongchang.com/api/v1/knowledge/claims/837555MCP
get_claim(id=837555)