New Release|Introducing AI Stats Gateway|Read the docs

New Release

Social

Explore

Docs Roadmap Tools Migrate Pricing Pricing Calculator Works With

More

Contributors DB Monitor Status Contact

Company

About Announcements Terms Privacy

Social

Explore

Docs Roadmap Tools Migrate Pricing Pricing Calculator Works With

More

Contributors DB Monitor Status Contact

Company

About Announcements Terms Privacy

© 2025 AI StatsIf you run into any issues or notice any data errors, please visit our GitHub and report an issue.

AI Stats

AI Stats

Home Organisations Models Benchmarks API Providers

Home Organisations Models Benchmarks API Providers

OSWorld - Benchmark Leaderboard & Model Performance | AI Stats

OSWorld

View benchmark source

Recorded Results

1

Average Score

0.66

Score Range

0.66 - 0.66

Leading Model

0.66 - Claude Opus 4.5

Scores Over Time

Individual benchmark scores plotted by date.

Models Using This Benchmark

Organisation	Model	Reported	Top Score	Info	Self Reported	Source
Anthropic	Claude Opus 4.5	24 Nov 2025	0.66	Pass@1; Avg@5, 64k Thinking	Yes	Source