AgentScout / Best lists / UI-TARS-desktop alternatives
UI-TARS-desktop alternatives (2026)
UI-TARS-desktop is a browser/computer use tool: ByteDance's open-source multimodal agent stack, including a desktop app and Agent TARS for controlling computers and browsers. Pricing, as the maker states it: Open source. Its GitHub repo has 39,297 stars. We found 74 tools in the directory that do overlapping work (browsing the web, controlling a computer, open-source agents); 52 of them are open source and 13 list a free or freemium plan. Below are the 10 closest, ranked by how many tasks and sub-tasks they share with UI-TARS-desktop, then by GitHub stars.
Updated
Quick comparison Top 10 of 11 listed below. Pricing as the maker states it; blank means we don't have it.
| Tool | Pricing | Best for | Open source |
|---|---|---|---|
 Ui.Vision RPA | Open source | Automate the browser | Yes |
 Terminator | Open source | Automate the browser | Yes |
 Anchor Action Index | Freemium | Write code | |
 Cua | Open source + cloud | Automate the browser | Yes |
 Midscene.js | Open source | Automate the browser | Yes |
 Nanobrowser | Open source | Automate the browser | Yes |
 RobotGo | Open source | Automate the browser | Yes |
 UFO | Open source | Automate the browser | Yes |
 Windows-MCP | Open source | Automate the browser | Yes |
 AstronRPA | Open source | Write code | Yes |
- Ui.Vision RPA (Open source): Browser and desktop automation extension with OCR and image recognition, plus an MCP server so AI agents can create and run macros. Consider it because it is ready to use without building anything; it also covers MCP servers and images and design.
- Terminator (Open source): Desktop automation library and MCP server that lets AI assistants control any Windows app through the accessibility tree, like Playwright for the desktop. Consider it because it also covers MCP servers and agent frameworks and SDKs.
- Anchor Action Index (Freemium): Shared memory of web workflows for Anchor's cloud browsers: computer-use agents replay proven paths as deterministic code and the model only steps in where a page has changed, exposed as tools on Anchor's MCP server and REST API. Consider it because it also covers memory and MCP servers.
- Cua (Open source + cloud): Open-source infrastructure for computer-use agents: sandboxes, drivers across operating systems and benchmarks. Consider it because it also covers sandboxes and secure execution and evals and benchmarks.
- Midscene.js (Open source): Vision-driven UI automation and E2E testing framework where a GUI agent carries out natural-language steps on web, mobile and desktop apps. Consider it because it also covers testing and QA and agent frameworks and SDKs.
- Nanobrowser (Open source): Open-source Chrome extension agent that automates web tasks in your own browser. Consider it because it is ready to use without building anything.
- RobotGo (Open source): Go library for cross-platform desktop automation and AI computer use that controls mouse, keyboard, screen, windows and processes. Consider it because it also covers agent frameworks and SDKs.
- UFO (Open source): Microsoft's desktop agent framework that operates Windows apps and orchestrates multi-device tasks as task graphs. Consider it because it also covers multi-agent orchestration and agent frameworks and SDKs.
- Windows-MCP (Open source): Lightweight MCP server for computer use on Windows that lets agents navigate files, control apps and interact with the UI. Consider it because it also covers MCP servers.
- AstronRPA (Open source): iFlytek's open-source enterprise RPA desktop app with a low-code visual designer whose workflows can be called by AI agents. Consider it because it is ready to use without building anything; it also covers workflow automation and images and design.
UI-TARS-desktop vs alternatives at a glance
| Tool | Pricing | Open source / licence | GitHub stars | Good for |
|---|---|---|---|---|
 UI-TARS-desktop | Open source | Apache License 2.0 | 39,297 | browsing the web, controlling a computer |
| Ui.Vision RPA | Open source | Yes | 2,083 | browsing the web, controlling a computer |
| Terminator | Open source | Yes | 1,656 | browsing the web, controlling a computer |
| Anchor Action Index | Freemium | No | — | writing code, browsing the web |
| Cua | Open source + cloud | MIT | 29,966 | browsing the web, controlling a computer |
| Midscene.js | Open source | MIT | 15,311 | browsing the web, controlling a computer |
| Nanobrowser | Open source | Apache License 2.0 | 14,108 | browsing the web, automating workflows |
| RobotGo | Open source | Apache License 2.0 | 10,871 | browsing the web, controlling a computer |
| UFO | Open source | MIT | 10,056 | browsing the web, controlling a computer |
| Windows-MCP | Open source | MIT | 9,404 | browsing the web, controlling a computer |
| AstronRPA | Open source | Yes | 5,276 | writing code, browsing the web |
More: UI-TARS-desktop listing · All browser/computer use tools · All alternatives pages
Related searches: AI computer-use agents · Free & open-source browser agents
Alternatives are picked from directory data only: same category plus shared tasks and sub-tasks. Pricing is as each maker states it; GitHub stars as recorded on each listing.