// BENCHMARK

OSWorld benchmark

AI model leaderboard for the OSWorld benchmark. Compare how large language models score on OSWorld, see the full ranking, and understand what this AI benchmark measures. Claude Opus 4.5 currently leads with 66.26. Computer-use benchmark — real desktop environment tasks across Linux/macOS, agents control GUI.

Leaderboard

# Model Organization Score Variant Source
#1 Claude Opus 4.5 Anthropic 66.26 p@1-avg@5 official ↗

Frequently asked questions about OSWorld

What is the OSWorld benchmark?

Computer-use benchmark — real desktop environment tasks across Linux/macOS, agents control GUI.

How is the OSWorld benchmark scored?

OSWorld is scored using the pass@1 metric, where a higher score is better. GenAIList aggregates reported scores from model providers and papers into a single ranked leaderboard.

Which AI model scores highest on OSWorld?

As of the latest reported scores on GenAIList, Claude Opus 4.5 achieves the highest result on OSWorld with a score of 66.26.

Is a higher OSWorld score better?

Yes. On OSWorld a higher score indicates better performance, so models near the top of the leaderboard are the strongest.