Physics Problem Solving on UGPhysics
67.6AccuracyGPT-5
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| GPT-5Reasoning Effort=High2025.08 | 67.6 | 4 | |
| o3Reasoning Effort=High2025.08 | 65.2 | 2.1 | |
| GPT-5Reasoning Effort=Low2025.08 | 63.6 | — | |
| o3Reasoning Effort=Low2025.08 | 63.1 | — | |
| o4-miniReasoning Effort=High2025.08 | 62.2 | 4.5 | |
| o3-miniReasoning Effort=High2025.08 | 60.7 | 4 | |
| o4-miniReasoning Effort=Low2025.08 | 57.7 | — | |
| o3-miniReasoning Effort=Low2025.08 | 56.7 | — | |
| Gemini-2.5-ProReasoning Effort=Low2025.08 | 56 | — | |
| Gemini-2.5-ProReasoning Effort=High2025.08 | 55.4 | -0.6 | |
| Claude-Sonnet-4Reasoning Effort=High2025.08 | 53.2 | 0.8 | |
| Claude-Sonnet-4Reasoning Effort=Low2025.08 | 52.4 | — |