Evaluation sources/OSWorld
ObservingProfessional work
OSWorld
XLANG Lab · Public model evaluation results from OSWorld.
Model results from this source
| GPT-5.6 Sol | 62.6 | 62.6 | 2026-07-09 |
| GPT-5.6 Terra | 50.2 | 50.2 | 2026-07-09 |
| GPT-5.6 Luna | 45.6 | 45.6 | 2026-07-09 |