xlang-ai/OSWorld

[NeurIPS 2024] OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments

[view on github]last commit: Oct 20, 2026

stars

3,225

7d

+23

30d

+73

90d

+201

## star history

## found in

Awesome Open Source AI/📈 9. Evaluation, Benchmarks & Datasets