metadata
license: apache-2.0
language:
- en
pipeline_tag: image-text-to-text
tags:
- multimodal
- gui
library_name: transformers
UI-TARS-7B-SFT
UI-TARS-2B-SFT | UI-TARS-7B-SFT | UI-TARS-7B-DPO(Recommended) | UI-TARS-72B-SFT | UI-TARS-72B-DPO(Recommended)
Introduction
UI-TARS is a next-generation native GUI agent model designed to interact seamlessly with graphical user interfaces (GUIs) using human-like perception, reasoning, and action capabilities. Unlike traditional modular frameworks, UI-TARS integrates all key components—perception, reasoning, grounding, and memory—within a single vision-language model (VLM), enabling end-to-end task automation without predefined workflows or manual rules.
This repository contains the model for the paper UI-TARS: Pioneering Automated GUI Interaction with Native Agents.
Code: https://github.com/bytedance/UI-TARS
Performance
Perception Capabilty Evaluation
| Model | VisualWebBench | WebSRC | SQAshort |
|---|---|---|---|
| Qwen2-VL-7B | 73.3 | 81.8 | 84.9 |
| Qwen-VL-Max | 74.1 | 91.1 | 78.6 |
| Gemini-1.5-Pro | 75.4 | 88.9 | 82.2 |
| UIX-Qwen2-7B | 75.9 | 82.9 | 78.8 |
| Claude-3.5-Sonnet | 78.2 | 90.4 | 83.1 |
| GPT-4o | 78.5 | 87.7 | 82.3 |
| UI-TARS-2B | 72.9 | 89.2 | 86.4 |
| UI-TARS-7B | 79.7 | 93.6 | 87.7 |
| UI-TARS-72B | 82.8 | 89.3 | 88.6 |
Grounding Capability Evaluation
- ScreenSpot Pro
| Agent Model | Dev-Text | Dev-Icon | Dev-Avg | Creative-Text | Creative-Icon | Creative-Avg | CAD-Text | CAD-Icon | CAD-Avg | Scientific-Text | Scientific-Icon | Scientific-Avg | Office-Text | Office-Icon | Office-Avg | OS-Text | OS-Icon | OS-Avg | Avg-Text | Avg-Icon | Avg |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| QwenVL-7B | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.7 | 0.0 | 0.4 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.1 | 0.0 | 0.1 |
| GPT-4o | 1.3 | 0.0 | 0.7 | 1.0 | 0.0 | 0.6 | 2.0 | 0.0 | 1.5 | 2.1 | 0.0 | 1.2 | 1.1 | 0.0 | 0.9 | 0.0 | 0.0 | 0.0 | 1.3 | 0.0 | 0.8 |
| SeeClick | 0.6 | 0.0 | 0.3 | 1.0 | 0.0 | 0.6 | 2.5 | 0.0 | 1.9 | 3.5 | 0.0 | 2.0 | 1.1 | 0.0 | 0.9 | 2.8 | 0.0 | 1.5 | 1.8 | 0.0 | 1.1 |
| Qwen2-VL-7B | 2.6 | 0.0 | 1.3 | 1.5 | 0.0 | 0.9 | 0.5 | 0.0 | 0.4 | 6.3 | 0.0 | 3.5 | 3.4 | 1.9 | 3.0 | 0.9 | 0.0 | 0.5 | 2.5 | 0.2 | 1.6 |
| OS-Atlas-4B | 7.1 | 0.0 | 3.7 | 3.0 | 1.4 | 2.3 | 2.0 | 0.0 | 1.5 | 9.0 | 5.5 | 7.5 | 5.1 | 3.8 | 4.8 | 5.6 | 0.0 | 3.1 | 5.0 | 1.7 | 3.7 |
| ShowUI-2B | 16.9 | 1.4 | 9.4 | 9.1 | 0.0 | 5.3 | 2.5 | 0.0 | 1.9 | 13.2 | 7.3 | 10.6 | 15.3 | 7.5 | 13.5 | 10.3 | 2.2 | 6.6 | 10.8 | 2.6 | 7.7 |
| CogAgent-18B | 14.9 | 0.7 | 8.0 | 9.6 | 0.0 | 5.6 | 7.1 | 3.1 | 6.1 | 22.2 | 1.8 | 13.4 | 13.0 | 0.0 | 10.0 | 5.6 | 0.0 | 3.1 | 12.0 | 0.8 | 7.7 |
| Aria-UI | 16.2 | 0.0 | 8.4 | 23.7 | 2.1 | 14.7 | 7.6 | 1.6 | 6.1 | 27.1 | 6.4 | 18.1 | 20.3 | 1.9 | 16.1 | 4.7 | 0.0 | 2.6 | 17.1 | 2.0 | 11.3 |
| UGround-7B | 26.6 | 2.1 | 14.7 | 27.3 | 2.8 | 17.0 | 14.2 | 1.6 | 11.1 | 31.9 | 2.7 | 19.3 | 31.6 | 11.3 | 27.0 | 17.8 | 0.0 | 9.7 | 25.0 | 2.8 | 16.5 |
| Claude Computer Use | 22.0 | 3.9 | 12.6 | 25.9 | 3.4 | 16.8 | 14.5 | 3.7 | 11.9 | 33.9 | 15.8 | 25.8 | 30.1 | 16.3 | 26.9 | 11.0 | 4.5 | 8.1 | 23.4 | 7.1 | 17.1 |
| OS-Atlas-7B | 33.1 | 1.4 | 17.7 | 28.8 | 2.8 | 17.9 | 12.2 | 4.7 | 10.3 | 37.5 | 7.3 | 24.4 | 33.9 | 5.7 | 27.4 | 27.1 | 4.5 | 16.8 | 28.1 | 4.0 | 18.9 |
| UGround-V1-7B | - | - | 35.5 | - | - | 27.8 | - | - | 13.5 | - | - | 38.8 | - | - | 48.8 | - | - | 26.1 | - | - | 31.1 |
| UI-TARS-2B | 47.4 | 4.1 | 26.4 | 42.9 | 6.3 | 27.6 | 17.8 | 4.7 | 14.6 | 56.9 | 17.3 | 39.8 | 50.3 | 17.0 | 42.6 | 21.5 | 5.6 | 14.3 | 39.6 | 8.4 | 27.7 |
| UI-TARS-7B | 58.4 | 12.4 | 36.1 | 50.0 | 9.1 | 32.8 | 20.8 | 9.4 | 18.0 | 63.9 | 31.8 | 50.0 | 63.3 | 20.8 | 53.5 | 30.8 | 16.9 | 24.5 | 47.8 | 16.2 | 35.7 |
| UI-TARS-72B | 63.0 | 17.3 | 40.8 | 57.1 | 15.4 | 39.6 | 18.8 | 12.5 | 17.2 | 64.6 | 20.9 | 45.7 | 63.3 | 26.4 | 54.8 | 42.1 | 15.7 | 30.1 | 50.9 | 17.5 | 38.1 |
- ScreenSpot v2
| Method | Mobile-Text | Mobile-Icon/Widget | Desktop-Text | Desktop-Icon/Widget | Web-Text | Web-Icon/Widget | Avg |
|---|---|---|---|---|---|---|---|
| Agent Framework | |||||||
| GPT-4o (SeeClick) | 85.2 | 58.8 | 79.9 | 37.1 | 72.7 | 30.1 | 63.6 |
| GPT-4o (OS-Atlas-4B) | 95.5 | 75.8 | 79.4 | 49.3 | 90.2 | 66.5 | 79.1 |
| GPT-4o (OS-Atlas-7B) | 96.2 | 83.4 | 89.7 | 69.3 | 94.0 | 79.8 | 87.1 |
| Agent Model | |||||||
| SeeClick | 78.4 | 50.7 | 70.1 | 29.3 | 55.2 | 32.5 | 55.1 |
| OS-Atlas-4B | 87.2 | 59.7 | 72.7 | 46.4 | 85.9 | 63.1 | 71.9 |
| OS-Atlas-7B | 95.2 | 75.8 | 90.7 | 63.6 | 90.6 | 77.3 | 84.1 |
| Our Model | |||||||
| UI-TARS-2B | 95.2 | 79.1 | 90.7 | 68.6 | 87.2 | 78.3 | 84.7 |
| UI-TARS-7B | 96.9 | 89.1 | 95.4 | 85.0 | 93.6 | 85.2 | 91.6 |
| UI-TARS-72B | 94.8 | 86.3 | 91.2 | 87.9 | 91.5 | 87.7 | 90.3 |
Online Agent Capability Evaluation
| Method | OSWorld (Online) | AndroidWorld (Online) |
|---|---|---|
| Agent Framework | ||
| GPT-4o (UGround) | - | 32.8 |
| GPT-4o (Aria-UI) | 15.2 | 44.8 |
| GPT-4o (Aguvis-7B) | 14.8 | 37.1 |
| GPT-4o (Aguvis-72B) | 17.0 | - |
| GPT-4o (OS-Atlas-7B) | 14.6 | - |
| Agent Model | ||
| GPT-4o | 5.0 | 34.5 (SoM) |
| Gemini-Pro-1.5 | 5.4 | 22.8 (SoM) |
| Aguvis-72B | 10.3 | 26.1 |
| Claude Computer-Use | 14.9 (15 steps) | 27.9 |
| Claude Computer-Use | 22.0 (50 steps) | - |
| Our Model | ||
| UI-TARS-7B-SFT | 17.7 (15 steps) | 33.0 |
| UI-TARS-7B-DPO | 18.7 (15 steps) | - |
| UI-TARS-72B-SFT | 18.8 (15 steps) | 46.6 |
| UI-TARS-72B-DPO | 22.7 (15 steps) | - |
| UI-TARS-72B-DPO | 24.6 (50 steps) | - |
Citation
If you find our paper and model useful in your research, feel free to give us a cite.
@article{uitars2025,
author = {Yujia Qin, Yining Ye, Junjie Fang, Haoming Wang, Shihao Liang, Shizuo Tian, Junda Zhang, Jiahao Li, Yunxin Li, Shijue Huang, Wanjun Zhong, Kuanye Li, Jiale Yang, Yu Miao, Woyu Lin, Longxiang Liu, Xu Jiang, Qianli Ma, Jingyu Li, Xiaojun Xiao, Kai Cai, Chuang Li, Yaowei Zheng, Chaolin Jin, Chen Li, Xiao Zhou, Minchao Wang, Haoli Chen, Zhaojian Li, Haihua Yang, Haifeng Liu, Feng Lin, Tao Peng, Xin Liu, Guang Shi},
title = {UI-TARS: Pioneering Automated GUI Interaction with Native Agents},
journal = {arXiv preprint arXiv:2501.12326},
url = {https://github.com/bytedance/UI-TARS},
year = {2025}
}