AA CritPt — leaderboard
Artificial Analysis independent evaluation of CritPt: 71 research-level physics reasoning tasks across 11 subfields.
Metric: Accuracy (%). Source: artificialanalysis.ai. Status: saturation imminent. 450 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.6 Sol (Max) | 32.29 |
| 2 | GPT-5.6 Terra (Max) | 30 |
| 3 | Claude Mythos 5 | 28.6 |
| 4 | GPT-5.6 Sol (xHigh) | 28.57 |
| 5 | GPT-5.5 (xHigh) | 27.14 |
| 6 | GPT-5.6 Terra (xHigh) | 27.14 |
| 7 | GPT-5.5 | 27.1 |
| 8 | GPT-5.6 Sol (High) | 25.71 |
| 9 | GPT-5.5 (High) | 25.43 |
| 10 | Kimi K3 | 23.43 |
| 11 | GPT-5.4 (xHigh) | 23.43 |
| 12 | GPT-5.6 Terra (High) | 22.86 |
| 13 | GPT-5.6 Sol (Medium) | 22.86 |
| 14 | Claude Opus 4.8 | 20.9 |
| 15 | GLM-5.2 (Max) | 20.86 |
Interactive version: theaggregate.ai/benchmark?slug=aa-critpt · How the rankings work · Data refreshed daily, snapshot 2026-07-22.