The Year in AI Papers
The year/Independent research

Paper 2509.02544

UI-TARS-2 Technical Report: Advancing GUI Agent with Multi-Turn Reinforcement Learning

Haoming Wang, Haoyang Zou, Huatong Song, Jiazhan Feng, Junjie Fang, Junting Lu, Longxiang Liu, Qinyu Luo, Shihao Liang, Shijue Huang, Wanjun Zhong, Yining Ye, et al.

Published
Sep 2025
Research lab
Independent
Citations
163
GitHub
11K stars

01 In brief

Summary

UI-TARS-2 is a native GUI-centered agent model developed by ByteDance Seed, designed to handle both structured computer-use tasks and dynamic game environments.

It addresses challenges in data scalability, multi-turn reinforcement learning (RL), GUI-only operation limits, and environment stability through a systematic methodology: a data flywheel for scalable data generation, a stabilized multi-turn RL framework, a hybrid GUI environment integrating file systems and terminals, and a unified sandbox platform for large-scale rollouts.

The model achieves significant improvements over its predecessor UI-TARS-1.5, reaching 88.2 on Online-Mind2Web, 47.5 on OSWorld, 50.6 on WindowsAgentArena, and 73.3 on AndroidWorld, outperforming strong baselines like Claude and OpenAI agents.

In game environments, it attains a mean normalized score of 59.8 across a 15-game suite, roughly 60% of human-level performance, and remains competitive with frontier models like OpenAI o3 on LMGame-Bench.

The model also generalizes to long-horizon information-seeking tasks and software engineering benchmarks, highlighting its robustness.

Detailed analyses of training dynamics provide insights into achieving stability and efficiency in large-scale agent RL, underscoring UI-TARS-2's potential to advance GUI agents and generalize to real-world interactive scenarios.

02 From the paper

Abstract

The development of autonomous agents for graphical user interfaces (GUIs) presents major challenges in artificial intelligence. While recent advances in native agent models have shown promise by unifying perception, reasoning, action, and memory through end-to-end learning, open problems remain in data scalability, multi-turn reinforcement learning (RL), the limitations of GUI-only operation, and environment stability. In this technical report, we present UI-TARS-2, a native GUI-centered agent model that addresses these challenges through a systematic training methodology: a data flywheel for scalable data generation, a stabilized multi-turn RL framework, a hybrid GUI environment that integrates file systems and terminals, and a unified sandbox platform for large-scale rollouts. Empirical evaluation demonstrates that UI-TARS-2 achieves significant improvements over its predecessor UI-TARS-1.5. On GUI benchmarks, it reaches 88.2 on Online-Mind2Web, 47.5 on OSWorld, 50.6 on WindowsAgentArena, and 73.3 on AndroidWorld, outperforming strong baselines such as Claude and OpenAI agents. In game environments, it attains a mean normalized score of 59.8 across a 15-game suite-roughly 60% of human-level performance-and remains competitive with frontier proprietary models (e.g., OpenAI o3) on LMGame-Bench. Additionally, the model can generalize to long-horizon information-seeking tasks and software engineering benchmarks, highlighting its robustness across diverse agent tasks. Detailed analyses of training dynamics further provide insights into achieving stability and efficiency in large-scale agent RL. These results underscore UI-TARS-2's potential to advance the state of GUI agents and exhibit strong generalization to real-world interactive scenarios.