Application
Agents & OrchestrationAutomation & Integration
UI-TARS Desktop
1A 2GUI 3agent 4that 5controls 6computers
the six Ws · specification
W1
Who
Maintained by ByteDance.
W2
What
A GUI agent that controls your computer visually.
W3
Where
A desktop app driving your screen and inputs.
W4
When
When automating desktop and browser GUIs.
W5
Why
Acts on any interface using vision, not brittle scripts.
W6
With
The desktop app and a compatible vision model.
W7
Watch
Vendor-maintained (ByteDance); open-source (Apache-2.0). The agent controls your mouse, keyboard, and screen, so supervise it; sends screenshots to the model you configure.