OSWorld 2 (Aug 8) Offline, 500-step Partial Score
Agents · 2026-08-08
OSWorld 2.0 v2026.08.08 offline 82-task subset, partial completion credit with 500 steps.
Top models (higher is better)
| Model | Score |
|---|---|
| Claude Opus 5 | 70.2 |
| GPT-5.6 Sol | 64.1 |
Agents · 2026-08-08
OSWorld 2.0 v2026.08.08 offline 82-task subset, partial completion credit with 500 steps.
| Model | Score |
|---|---|
| Claude Opus 5 | 70.2 |
| GPT-5.6 Sol | 64.1 |