| # | Model | Frontier | Math | Cap. | Cost |
|---|
| 1 | | 52.4% | 100.0 | 91.8 | $67.50/M |
| 2 | | 51.7% | 100.0 | 89.6 | $11.25/M |
| 3 | | 50.0% | — | 89.1 | $67.50/M |
| 4 | | 47.6% | 95.3 | 87.4 | $5.63/M |
| 5 | | 47.2% | 98.3 | 89.2 | $10.00/M |
| 6 | | 43.8% | 97.8 | 87.4 | $10.00/M |
| 7 | | 40.7% | — | 86.8 | $10.00/M |
| 7 | | 40.7% | 96.1 | 85.1 | $4.81/M |
| 8 | | 39.0% | 95.6 | 86.2 | $3.38/M |
| 8 | | 39.0% | 96.1 | 82.9 | $1.36/M |
| 9 | | 37.6% | 91.4 | 84.8 | — |
| 10 | | 36.9% | 95.6 | 86.4 | $4.50/M |
| 11 | | 35.6% | 92.8 | 82.4 | $1.13/M |
| 12 | | 33.4% | 92.2 | 82.3 | $1.50/M |
| 13 | | 32.4% | 91.4 | 81.8 | $3.44/M |
| 14 | | 31.0% | 88.6 | 81.6 | $3.44/M |
| 15 | | 28.3% | 87.2 | 80.8 | $1.69/M |
| 16 | | 27.2% | 86.7 | 78.0 | $0.69/M |
| 17 | | 26.2% | 90.6 | 81.1 | $0.73/M |
| 18 | | 25.9% | 87.8 | 78.7 | $0.46/M |
| 19 | | 24.8% | 81.7 | 78.6 | $1.93/M |
| 20 | | 20.7% | 48.1 | 81.4 | $10.00/M |
| 21 | | 18.7% | 83.9 | 79.2 | $3.50/M |
| 22 | | 16.4% | 80.0 | 78.6 | $0.93/M |
| 23 | | 14.1% | 84.2 | 78.3 | $3.44/M |
| 24 | | 12.4% | 76.9 | 73.9 | $1.93/M |
| 25 |  Claude Sonnet 4.5 Anthropic | 9.3% | 35.6 | 78.9 | $6.00/M |
| 25 | | 9.3% | 73.3 | 75.2 | $26.25/M |
| 26 | | 8.3% | 81.1 | 73.4 | $0.14/M |
| 27 | | 5.9% | 40.0 | 76.5 | $30.00/M |
| 28 | | 5.5% | 38.3 | 70.4 | $3.50/M |
| 29 | | 4.8% | — | 72.9 | $0.85/M |
| 30 | | 4.5% | 42.2 | 75.4 | $30.00/M |
| 30 | | 4.5% | 44.7 | 68.9 | $0.70/M |
| 31 | Claude Haiku 4.5 Anthropic | 4.1% | 35.8 | 75.3 | $2.00/M |
| 31 | | 4.1% | 28.9 | 74.7 | $6.00/M |
| 32 | Claude 3.5 Sonnet Anthropic | 2.1% | 8.5 | 67.5 | — |
| 33 | | 1.7% | 17.8 | 63.1 | $0.39/M |
| 33 | | 1.7% | 15.8 | 78.7 | $0.35/M |
| 34 | | 1.0% | 28.9 | 64.4 | $0.17/M |
| 34 | | 1.0% | 16.1 | 66.7 | — |
| 35 | Claude 3.5 Haiku Anthropic | 0.3% | 4.3 | 61.4 | $1.60/M |
| 35 | | 0.3% | 6.4 | 63.0 | $4.38/M |
| 35 |  Mistral Large 2 (Nov 2024) Mistral | 0.3% | 7.8 | 62.5 | — |
44 of 44 models, ranked by FrontierMath
