| # | Model | SimpleQA | Reasoning | Cap. | Cost |
|---|
| 1 | | 77.3% | 94.1 | 86.4 | $4.50/M |
| 2 | | 72.9% | 92.6 | 84.8 | — |
| 3 | | 68.4% | 92.8 | 86.2 | $3.38/M |
| 4 | | 68.3% | — | 91.6 | $20.00/M |
| 5 | | 67.5% | 72.6 | 76.5 | $1.56/M |
| 6 | | 67.4% | 83.2 | 82.4 | $1.13/M |
| 7 | | 64.5% | 93.9 | 91.8 | $67.50/M |
| 8 | | 63.1% | 94.0 | 89.6 | $11.25/M |
| 9 | | 58.5% | 91.6 | 84.7 | $2.21/M |
| 10 | | 57.0% | 89.6 | 81.6 | $0.54/M |
| 11 | | 56.0% | 85.3 | 78.3 | $3.44/M |
| 12 | | 53.5% | 93.4 | 88.5 | $3.00/M |
| 13 | | 53.0% | 81.8 | 79.2 | $3.50/M |
| 14 | | 50.6% | 90.2 | 87.4 | $10.00/M |
| 14 | | 50.6% | 86.2 | 81.8 | $3.44/M |
| 15 | | 49.1% | 87.4 | 81.1 | $0.73/M |
| 16 | | 48.9% | 87.6 | 81.6 | $3.44/M |
| 17 | | 47.8% | 94.6 | 89.1 | $67.50/M |
| 18 | | 44.8% | 93.3 | 87.4 | $5.63/M |
| 19 |  Inkling Thinking Machines Lab | 43.9% | 87.2 | — | $1.76/M |
| 20 | | 43.1% | — | 86.8 | $10.00/M |
| 21 | | 39.5% | 91.0 | 89.2 | $10.00/M |
| 22 | | 39.2% | 89.5 | 81.8 | $1.55/M |
| 23 | | 38.9% | 91.4 | 85.1 | $4.81/M |
| 24 | | 38.7% | 90.8 | 82.9 | $1.37/M |
| 25 | | 38.1% | 91.9 | 83.4 | $1.19/M |
| 26 | | 38.0% | 88.8 | 80.8 | $1.56/M |
| 27 | | 37.3% | 85.5 | 82.3 | $1.48/M |
| 28 | | 31.5% | 83.3 | 76.1 | $0.74/M |
| 29 | | 28.6% | 83.6 | 80.8 | $1.69/M |
| 30 |  DeepSeek R1 (0528) DeepSeek | 27.4% | 76.3 | 74.5 | $0.91/M |
| 31 | | 25.0% | 90.5 | 80.1 | $4.00/M |
| 32 | | 23.9% | 79.6 | 78.6 | $1.93/M |
| 33 | | 21.0% | 75.0 | 78.0 | $0.69/M |
| 34 |  Claude Sonnet 4.5 Anthropic | 13.0% | 73.7 | 78.9 | $6.00/M |
| 35 | | 12.2% | 69.4 | 73.4 | $0.14/M |
| 36 | | 12.0% | 78.5 | 78.7 | $0.46/M |
| 37 | | 9.6% | — | — | $0.18/M |
| 38 | Claude 3.5 Haiku Anthropic | 6.7% | 38.1 | 61.4 | $1.60/M |
39 of 39 models, ranked by SimpleQA
