OSWorld2.0: Benchmarking Computer Use Agents on Long-Horizon Real-World Tasks

• URL: http://arxiv.org/abs/2606.29537v1
• Date: Sun, 28 Jun 2026 17:59:17 GMT
• Title: OSWorld2.0: Benchmarking Computer Use Agents on Long-Horizon Real-World Tasks
• Abstract summary: OSWorld 2.0 is a benchmark of 108 long-horizon computer-use across everyday and professional tasks.<n>Each task represents a realistic end-to-end workflow that takes human users a median of about 1.6 hours to complete.
• Score: 98.07146747892239
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Existing computer-use benchmarks fail to capture the realism, complexity, and long-horizon demands of real-world computer use, limiting their ability to reveal the limitations of frontier agents. We introduce OSWorld 2.0, a benchmark of 108 long-horizon computer-use workflows across everyday and professional tasks, designed to capture complex and challenging real-world phenomena. Each task represents a realistic end-to-end workflow that takes human users a median of about 1.6 hours to complete and requires an average of 318 tool calls with Claude Opus 4.7 using maximum thinking, compared with about 30 in OSWorld 1.0. OSWorld 2.0 targets challenge phenomena that are common in real workflows yet underrepresented in prior benchmarks, spanning interaction-design challenges such as streaming interaction and dynamic environments, as well as agent-pattern challenges such as cross-source reasoning, implicit-state inference, and visual-spatial precision. Tasks are grounded in authentic input artifacts and cross-referenced against realistic stateful user profile data, and include separate safety reports auditing safety-sensitive execution. Under our primary binary-completion metric at 500 steps, Claude Opus 4.8 with maximum thinking and batched tool calls scores best but still completes only 20.6% of tasks at a 54.8% partial score; GPT-5.5 is far more token-efficient yet plateaus near 13%. These results show that current agents are still far from professional-level computer use: rather than stumbling on basic GUI control or coding, they lose track of constraints, miss information that arrives mid-task, guess rather than ask the user, and skip verification, struggling most when a task hinges on hidden state they must recover.

Related papers

• Running the Gauntlet: Re-evaluating the Capabilities of Agents Beyond Familiar Environments [67.2234316079859]
  GauntletBench is a web-based benchmark for evaluating agent generalisation in challenging scenarios.<n>It focuses on three underexplored capabilities (temporal perception, graphical understanding, and 3D reasoning) across five less-covered professional applications.<n>Our empirical results reveal that frontier agentic systems remain far from achieving human-level performance.
  arXiv  Detail & Related papers  (2026-06-12T12:32:24Z)
• SaaS-Bench: Can Computer-Use Agents Leverage Real-World SaaS to Solve Professional Workflows? [29.5539656241093]
  Computer-Using Agents (CUAs) are rapidly extending large language models (LLMs) beyond text-based reasoning toward action execution in more complex environments.<n>Existing benchmarks often rely on simplified settings, isolated tasks, or short-horizon interactions.<n>We introduce a benchmark built on 23 deployable systems across six professional domains, containing 106 tasks grounded in realistic work scenarios.<n>Experiments show that representative LLM-based agents struggle on-Bench, with even the strongest model completing fewer than 4% of tasks end-to-end.
  arXiv  Detail & Related papers  (2026-05-15T09:35:15Z)
• AcademiClaw: When Students Set Challenges for AI Agents [69.70303995121154]
  AcademiClaw is a benchmark of 80 complex, long-horizon tasks sourced directly from university students' real academic disconnects.<n>We show that even the best of models achieves only a 55% pass rate.<n>We hope that AcademiClaw and its open-sourced data and code can serve as a useful resource for the OpenClaw community.
  arXiv  Detail & Related papers  (2026-05-04T14:40:42Z)
• WindowsWorld: A Process-Centric Benchmark of Autonomous GUI Agents in Professional Cross-Application Environments [34.06270058972]
  WindowsWorld is designed to assess GUI Agents on complex multi-step tasks that mirror real-world professional activities.<n>Our methodology uses a multi-agent framework steered by 16 occupations to generate four difficulty-level tasks.<n>The resulting benchmark contains 181 tasks with an average of 5.0 sub-goals across 17 common desktop applications.
  arXiv  Detail & Related papers  (2026-04-30T12:13:27Z)
• OccuBench: Evaluating AI Agents on Real-World Professional Tasks via Language Environment Simulation [57.505743202759646]
  OccuBench is a benchmark covering 100 real-world professional task scenarios across 10 industry categories and 65 specialized domains.<n>Our multi-agent synthesis pipeline automatically produces evaluation instances with guaranteed solvability, calibrated difficulty, and document-grounded diversity.
  arXiv  Detail & Related papers  (2026-04-13T00:27:32Z)
• World of Workflows: a Benchmark for Bringing World Models to Enterprise Systems [6.643798239990755]
  Frontier large language models (LLMs) excel in many domains, yet they remain untested in complex enterprise systems.<n>We introduce World of Warcraft (WoW), a realistic ServiceNow-based environment incorporating 4,000+ business rules and 55 active embedded in the system.<n>We reveal two major takeaways: (1) Frontier LLMs suffer from dynamics blindness, consistently failing to predict the invisible, cascading side effects of their actions, which leads to silent constraint violations, and (2) reliability in opaque systems requires grounded world modeling, where agents must mentally simulate hidden state transitions to bridge the observability gap when high-fidelity feedback is unavailable.
  arXiv  Detail & Related papers  (2026-01-29T18:51:54Z)
• Terminal-Bench: Benchmarking Agents on Hard, Realistic Tasks in Command Line Interfaces [126.23612941699565]
  Terminal-Bench 2.0 is a benchmark composed of 89 tasks in computer terminal environments inspired by problems from real world.<n>We show that frontier models and agents score less than 65% on the benchmark.<n>We publish the dataset and evaluation harness to assist developers and researchers in future work at https://www.tbench.ai/.
  arXiv  Detail & Related papers  (2026-01-17T01:29:30Z)
• UI-TARS-2 Technical Report: Advancing GUI Agent with Multi-Turn Reinforcement Learning [155.51875080423883]
  The development of autonomous agents for graphical user interfaces presents major challenges in artificial intelligence.<n>We present UI-TARS-2, a native GUI-centered agent model that addresses these challenges through a systematic training methodology.<n> Empirical evaluation demonstrates that UI-TARS-2 achieves significant improvements over its predecessor UI-TARS-1.5.
  arXiv  Detail & Related papers  (2025-09-02T17:44:45Z)
• UI-Vision: A Desktop-centric GUI Benchmark for Visual Perception and Interaction [16.731754927372585]
  We introduce UI-Vision, the first comprehensive, license-permissive benchmark for offline, fine-grained evaluation of computer use agents.<n>Unlike online benchmarks, UI-Vision provides dense, high-quality annotations of human demonstrations.<n>Our evaluation reveals critical limitations in state-of-the-art models like UI-TARS-72B.
  arXiv  Detail & Related papers  (2025-03-19T19:26:17Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.