OSUniverse: Benchmark for Multimodal GUI-navigation AI Agents

TL;DR

OSUniverse is a comprehensive benchmark designed to evaluate multimodal GUI-navigation AI agents across complex tasks, with automated validation and scalable difficulty, facilitating progress measurement in AI capabilities.

Contribution

It introduces a new, extensible benchmark with automated validation for assessing multimodal GUI-navigation AI agents' performance on complex desktop tasks.

Findings

State-of-the-art agents score below 50% on benchmark tasks

Automated validation achieves less than 2% error rate

Benchmark covers a range of task complexities from simple to multiapplication

Abstract

In this paper, we introduce OSUniverse: a benchmark of complex, multimodal desktop-oriented tasks for advanced GUI-navigation AI agents that focuses on ease of use, extensibility, comprehensive coverage of test cases, and automated validation. We divide the tasks in increasing levels of complexity, from basic precision clicking to multistep, multiapplication tests requiring dexterity, precision, and clear thinking from the agent. In version one of the benchmark, presented here, we have calibrated the complexity of the benchmark test cases to ensure that the SOTA (State of the Art) agents (at the time of publication) do not achieve results higher than 50%, while the average white collar worker can perform all these tasks with perfect accuracy. The benchmark can be scored manually, but we also introduce an automated validation mechanism that has an average error rate less than 2%. Therefore, this benchmark presents solid ground for fully automated measuring of progress, capabilities and the effectiveness of GUI-navigation AI agents over the short and medium-term horizon. The source code of the benchmark is available at https://github.com/agentsea/osuniverse.