Artificial intelligence interface that manages your computer
Developed by ByteDance, UI-TARS is an open source agent infrastructure that integrates multimodal AI models with desktop interfaces. This system enables the creation of agents (AI agents) that autonomously perform tasks on the computer by processing visual data.
Updates
- August 2, 2026: Stars 36,779 → 38,404, latest release v0.3.0 (November 4, 2025).
What you get
- Autonomously perform desktop and browser tasks
- Human-like workflow by processing visual data
- Full integration with multi-modal AI models
Installation
npm install @agent-tars/cli@latest -gnpx @agent-tars/cli@latestRunning it
agent-tars --provider volcengine --model doubao-1-5-thinking-vision-pro-250428 --apiKey your-api-keyIf you don't write code
Describe a complex task you want me to perform on my computer or browser. For example, I might ask you to go to a specific website and enter data or manage a flight booking process. Please analyze the visual interface and complete the steps autonomously.
Related dictionary terms
Links
TreScout did not build this tool · we found it in GitHub trends and wrote it up. This page describes the repository as of 2026-06-18: The star count and our text belong to that day, the repository may have changed since. Check the repository link for the current state. This page was machine-translated from the Turkish original · the Turkish version prevails.