OSWorld-Verified — leaderboard
OSWorld-Verified evaluates model capability on agentic tasks from the linked upstream source with Score as the primary reported metric.
Metric: Score (self-reported). Source: benchmarklist.com. Status: saturation imminent. 20 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Claude Fable 5 (Max) | 85 |
| 2 | Claude Opus 4.8 | 83.4 |
| 3 | Claude Opus 4.8 (Max) | 83.4 |
| 4 | Claude Opus 4.7 | 82.8 |
| 5 | Claude Sonnet 5 | 81.2 |
| 6 | Claude Mythos Preview | 79.6 |
| 7 | GPT-5.5 | 78.7 |
| 8 | Claude Sonnet 4.6 | 78.5 |
| 9 | Gemini 3.5 Flash | 78.4 |
| 10 | Gemini 3.1 Pro (Preview) | 76.2 |
| 11 | GPT-5.4 | 75 |
| 12 | Qwen 3.7 Plus | 73.3 |
| 13 | Claude Opus 4.6 (Max) | 72.7 |
| 14 | MiniMax-M3 | 70.06 |
| 15 | Qwen 3.6 Plus | 62.5 |
Interactive version: theaggregate.ai/benchmark?slug=osworld-verified · How the rankings work · Data refreshed daily, snapshot 2026-07-22.