Reasoning on GSM8K (p_T/p_S Evaluation)
97.1p_T ScoreO3
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| O32025.10 | 97.1 | 1 | 93.6 | 1 | -3.5 | 95.5 | 92.3 | 1 | |
| GPT-5-minisize=mini2025.10 | 96.9 | 1 | 80.1 | 7 | -16.8 | 71.4 | 73.3 | 6 | |
| Kimi-K22025.10 | 96.8 | 1 | 81.6 | 6 | -15.2 | 92.5 | 79.7 | 4 | |
| GPT-52025.10 | 96.8 | 1 | 82 | 6 | -14.7 | 81.4 | 77.5 | 5 | |
| O12025.10 | 96.5 | 2 | 82.7 | 5 | -13.8 | 97 | 82 | 3 | |
| O4-minisize=mini2025.10 | 96.5 | 2 | 69.4 | 11 | -27.1 | 94.9 | 68.4 | 9 | |
| GPT-5-chatmode=chat2025.10 | 96.4 | 2 | 56.2 | 14 | -40.2 | 85.2 | 53.7 | 13 | |
| DeepSeek-V32025.10 | 96.3 | 2 | 75.3 | 9 | -21 | 69.8 | 68.4 | 9 | |
| Llama-3.3-Instparameters=70B2025.10 | 96.2 | 2 | 89.5 | 3 | -6.7 | 51.9 | 76.6 | 6 | |
| DeepSeek-R12025.10 | 96.1 | 2 | 79.2 | 8 | -16.8 | 75.8 | 73.5 | 6 | |
| O3-minisize=mini2025.10 | 96.1 | 2 | 91.6 | 2 | -4.5 | 29.4 | 72.2 | 7 | |
| GPT-4o2025.10 | 95.9 | 3 | 84.2 | 4 | -11.7 | 48.4 | 71.2 | 8 | |
| Qwen2.5-Instparameters=72B2025.10 | 95.7 | 3 | 63.8 | 12 | -31.9 | 56.7 | 55.5 | 12 | |
| GPT-4.1-minisize=mini, version=4.12025.10 | 95.5 | 3 | 70.4 | 10 | -25.1 | 77.8 | 65.7 | 10 | |
| GPT-42025.10 | 95.1 | 4 | 88.3 | 3 | -6.8 | 83.1 | 84.3 | 2 | |
| GPT-4.1version=4.12025.10 | 95.1 | 4 | 70.3 | 10 | -24.8 | 90.1 | 68.2 | 9 | |
| GPT-5-nanosize=nano2025.10 | 94.8 | 4 | 84.4 | 4 | -10.4 | 15.5 | 63 | 11 | |
| Phi-42025.10 | 94.2 | 5 | 84.5 | 4 | -9.7 | 35.2 | 68.1 | 9 | |
| O1-minisize=mini2025.10 | 93.3 | 6 | 59.6 | 13 | -33.7 | 78.7 | 55.8 | 12 | |
| Ministral-Instparameters=8B, version=24102025.10 | 86.1 | 7 | 55.9 | 14 | -30.2 | 63.8 | 49.8 | 14 | |
| GPT-3.52025.10 | 76.4 | 7 | 40.2 | 15 | -36.2 | 50 | 34.5 | 15 |