OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments

TL;DR

OSWorld introduces a scalable, real computer environment benchmark for evaluating multimodal agents on diverse, real-world tasks across multiple operating systems, revealing significant gaps in current AI agent capabilities.

Contribution

We present OSWorld, the first scalable, real computer environment benchmark for multimodal agents, enabling evaluation of open-ended tasks across various OS and applications.

Findings

State-of-the-art models achieve only 12.24% success on OSWorld tasks.

Humans accomplish over 72.36% of tasks, highlighting AI limitations.

OSWorld reveals critical challenges in GUI grounding and operational knowledge for AI agents.

Abstract

Autonomous agents that accomplish complex computer tasks with minimal human interventions have the potential to transform human-computer interaction, significantly enhancing accessibility and productivity. However, existing benchmarks either lack an interactive environment or are limited to environments specific to certain applications or domains, failing to reflect the diverse and complex nature of real-world computer use, thereby limiting the scope of tasks and agent scalability. To address this issue, we introduce OSWorld, the first-of-its-kind scalable, real computer environment for multimodal agents, supporting task setup, execution-based evaluation, and interactive learning across various operating systems such as Ubuntu, Windows, and macOS. OSWorld can serve as a unified, integrated computer environment for assessing open-ended computer tasks that involve arbitrary applications. Building upon OSWorld, we create a benchmark of 369 computer tasks involving real web and desktop apps in open domains, OS file I/O, and workflows spanning multiple applications. Each task example is derived from real-world computer use cases and includes a detailed initial state setup configuration and a custom execution-based evaluation script for reliable, reproducible evaluation. Extensive evaluation of state-of-the-art LLM/VLM-based agents on OSWorld reveals significant deficiencies in their ability to serve as computer assistants. While humans can accomplish over 72.36% of the tasks, the best model achieves only 12.24% success, primarily struggling with GUI grounding and operational knowledge. Comprehensive analysis using OSWorld provides valuable insights for developing multimodal generalist agents that were not possible with previous benchmarks. Our code, environment, baseline models, and data are publicly available at https://os-world.github.io.