GH Repository · bytedance
UI-TARS-desktop
The Open-Source Multimodal AI Agent Stack: Connecting Cutting-Edge AI Models and Agent Infra
- stars
- 39,017
- 30-day movement
- +3913/day
- Related entries
- 67
- Connections
- 1
typescriptagentTypeScriptvlmvisioncoworkcomputer-usegui-operatornode/pnpmmcp-servertarsbrowser-usenodemcpgui-agentmultimodalui-tarsagent-tars
UI-TARS-desktop is ByteDance's open-source multimodal AI agent stack, described as connecting cutting-edge AI models with agent infrastructure. Based on its topics, it operates as an MCP server and GUI agent that uses vision-language models for computer use, browser use, and GUI operation.
Reach for it when you want a vision-driven agent that can control desktop and browser interfaces through MCP.
Use it to
- Run a GUI-operating agent on your desktop
- Expose computer-use capabilities via MCP server
- Automate browser interactions with a VLM agent
- Build on the open-source agent stack in TypeScript
For Developers building multimodal computer-use agents
- Role
- mcp-server
- Language
- TypeScript
- Licence
- Apache-2.0
- Forks
- 3,948
- Open issues
- 329
- Last push
- 2026-09-11
- Latest release
- v0.0.1 · 2025-01-22
topicscomputer-usegui-agentmcp-servermultimodalvlmbrowser-use