ContextQMD
Back to UI-TARS-desktop

Quick Start

docs/quick-start.md

0.3.06.6 KB
Original Source

Quick Start

We're excited to announce the support for UI-TARS-1.5! ๐ŸŽ‰๐ŸŽ‰๐ŸŽ‰

The previous version of UI-TARS Desktop version 0.0.8 will be upgraded to a new Desktop App 0.1.0 with support for both Computer and Browser operator.

Prerequisites

Please install Chrome (stable/beta/dev/canary), Edge (stable/beta/dev/canary), or Firefox (stable/beta/dev/nightly) for Browser Operator.

UI-TARS-desktop is currently only available for single monitor setup. Multi-monitor configuration may cause failure for some tasks.

Download

You can download the latest release version of UI-TARS Desktop from our releases page.

Note: If you have Homebrew installed, you can install UI-TARS Desktop by running the following command:

bash
brew install --cask ui-tars

Install

MacOS

  1. Drag UI TARS application into the Applications folder

  2. Enable the permission of UI TARS in MacOS:

  • System Settings -> Privacy & Security -> Accessibility
  • System Settings -> Privacy & Security -> Screen Recording
  1. Then open UI TARS application, you can see the following interface:

Windows

Still to run the application, you can see the following interface:

Run remote operator

The Remote Operator service will be discontinued on August 20, 2025. If you wish to deploy your own Remote Computer and Browser Agent after the free trial, you can explore Volcano Engine's OS Agent Services.

Deployment Links (in Chinese): Computer Use Agent and Browser Use Agent

Get model and run local operator

UI-TARS-1.5 on Hugging Face

  1. Click the button Deploy from Hugging Face on the top right corner of the page

  2. Select the model UI-TARS-1.5-7B

  3. Refer to README_deploy.md for detailed deployment instructions to obtain the Base URL, API Key, and Model Name.

  4. Open the UI-TARS Desktop App Settings and configure:

yaml
Language: en
VLM Provider: Hugging Face for UI-TARS-1.5
VLM Base URL: https:xxx
VLM API KEY: your_api_key
VLM Model Name: xxx

[!NOTE]

  1. For VLM Provider, make sure to select "Hugging Face for UI-TARS-1.5" to ensure proper VLM Action parsing.

  2. For VLM Base URL & VLM Model Name, you can checkout your huggingface endpoint page to see detail information. Please make sure Base URL ends with '/v1/'

  1. Click button starting a new chat

  2. Input the command to start a round of GUI operation tasks!

Doubao-1.5-UI-TARS on VolcEngine

  1. Visit the VolcEngine Doubao-1.5-UI-TARS page

  2. Click the button Try (็ซ‹ๅณไฝ“้ชŒ) on the top right corner of the page

  3. Click the API inference (API ๆŽฅๅ…ฅ) link

  4. Get your API Key from STEP 1 in the drawer panel.

  5. In STEP 2, authenticate your user info and switch to the OpenAI SDK tab to obtain Base Url and Model name๏ผš

  6. Open the UI-TARS Desktop App Settings and configure:

yaml
Language: cn
VLM Provider: VolcEngine Ark for Doubao-1.5-UI-TARS
VLM Base URL: https://ark.cn-beijing.volces.com/api/v3
VLM API KEY: YOUR_API_KEY
VLM Model Name: doubao-1.5-ui-tars-250328

[!NOTE] For VLM Provider, make sure to select "VolcEngine Ark for Doubao-1.5-UI-TARS" to ensure proper VLM Action parsing.

  1. Select the desired usage scenario before starting a new chat

[!NOTE] Before using Browser Operator mode, please ensure that Chrome, Edge, or Firefox is installed on your device.

  1. Input the command to start a round of GUI operation tasks!

More

At this point, you should have successfully launched the UI-TARS-Desktop App! To get the most out of UI-TARS and ensure stable usage, we recommend reviewing the following documentation: