Agentic computer use

OSWorld-Verified

Can the AI actually operate a computer — clicking, typing, and using real apps — to finish tasks on its own? Higher is better.

Rankings

Higher is better

OSWorld-Verified — frequently asked questions

What is OSWorld-Verified?
Can the AI actually operate a computer — clicking, typing, and using real apps — to finish tasks on its own? Higher is better.
Which AI model scores highest on OSWorld-Verified?
Qwen3.8-Max by Qwen holds the best OSWorld-Verified result among tracked models, at 86.1% (released Aug 3 2026). Higher scores are better on this benchmark.
What are the top 5 models on OSWorld-Verified?
1. Qwen3.8-Max (Qwen) — 86.1%; 2. Claude Fable 5 (Anthropic) — 85%; 3. Claude Opus 4.8 (Anthropic) — 83.4%; 4. Gemini 3.6 Flash (Google) — 83%; 5. Claude Sonnet 5 (Anthropic) — 81.2%.
How many models have a published OSWorld-Verified score?
16 tracked models have a published OSWorld-Verified score. Scores are the figures reported by each lab at that model's release, so this page is a record of results over time rather than a re-run leaderboard.
What is the best open model on OSWorld-Verified?
Qwen3.5 by Qwen is the highest-ranked model with downloadable weights on OSWorld-Verified, scoring 62.2% at rank 15 overall.
← All benchmarks