UI-TARS 7B
UI-TARS 7B is a multimodal vision-language model developed by ByteDance. It is specifically designed to navigate and interact with graphical user interfaces across desktop, web, mobile, and gaming environments.
UI-TARS 7B specs and pricing
- Provider: ByteDance
- Released: July 22, 2025
- Context window: 128K tokens
- Max output: 2K tokens
- Input: images, text
- Output: text
- Price: $0.10 input / $0.20 output per 1M tokens
What UI-TARS 7B is good at
- Automating tasks within desktop applications
- Navigating web browser interfaces
- Interacting with mobile system menus
To-dos to try with UI-TARS 7B
- Analyze this screenshot of a settings menu and identify the path to the privacy controls
- Navigate to the search bar on this webpage and enter the query for recent news
- Observe this game interface and describe the current health and inventory status
FAQ
- What is the primary purpose of UI-TARS 7B? It is designed to act as a vision-language agent capable of understanding and operating within GUI-based environments like browsers and mobile systems.
- Can this model process visual information? Yes, the model accepts both image and text inputs to interpret and interact with graphical interfaces.
- How much does UI-TARS 7B cost? UI-TARS 7B costs $0.10 per million input tokens and $0.20 per million output tokens through OpenRouter. In todo.is a typical agent step costs about 0.31 credits ($1 = 100 credits).
- What is the context window of UI-TARS 7B ? UI-TARS 7B can read up to 128K tokens at once and write up to 2K tokens in one reply.
- Can I use UI-TARS 7B without coding? Yes. In todo.is you pick UI-TARS 7B as your agent's model and write to-dos in plain words, on the web, the Android app, WhatsApp or Telegram. No API key needed: credits are included, or you can bring your own key.
JavaScript is required to use the todo.is app.