Physics Reasoning on PhysReason
73.3Overall ScoreARYA
Evaluation Results
| Method | Links | |
|---|---|---|
| ARYAPrompting Strategy=Zero-shot prompting, Parameters=02026.03 | 73.3 | |
| DeepSeek-R1Context=Best Published Baseline2026.03 | 56.8 | |
| Claude Opus 4.6Prompting Strategy=Zero-shot prompting2026.03 | 21.8 | |
| GPT-5.2Prompting Strategy=Zero-shot prompting2026.03 | 6.5 |