Skip to main content
Benchmark Registry
ModelsBenchmarksOrganizations

OSWorld

Benchmark
OSWorld
Evaluated by
XLang Lab
Release date
July 28, 2025
Version
OSWorld Verified
Metric
OSWorld-Verified success rate

Results

3 results

LatestHistory
All providersAlibaba (Tongyi)AnthropicGoogle (DeepMind)Meta AI (originally Facebook AI Research)Moonshot AIOpenAI
OSWorld Verified results
ProviderModelScoreSourceRegistry No.
OpenAIGPT-5.5 (xhigh)78.7%Source for GPT-5.5 (xhigh) on OSWorld Verified (opens in a new tab)10008
OpenAIGPT-5.4 (xhigh)75.0%Source for GPT-5.4 (xhigh) on OSWorld Verified (opens in a new tab)10007
OpenAIGPT-5.3-Codex (xhigh)64.7%Source for GPT-5.3-Codex (xhigh) on OSWorld Verified (opens in a new tab)10006
Page 1 of 1

© 2026 Densa Labs

Toggle between the data update date and the application build time.
Legal
Color theme