Ranking
Models with a blended list price under $0.15 per million tokens (3 input : 1 output), ordered by verification score.
| # | Model | Score | Cost / question | Cost / correct answer | Speed (tok/s) |
|---|---|---|---|---|---|
| 1 |  Mercury 2.5Inception | 58.2% | $0.00011 | $0.00019 | 157 |
| 2 |  gpt-oss-20bOpenAI | 51.2% | $0.000066 | $0.00013 | 103 |
| 3 |  Nemotron 3.5 LightningNVIDIA | 43.8% | $0.00049 | $0.0011 | 116 |
| 4 |  Ling 3.1 FlashinclusionAI | 41.2% | $0.00061 | $0.0015 | 7 |
| 5 |  Phi 4Microsoft | 41.2% | $0.00012 | $0.00029 | 103 |
| 6 |  Solar Mini 4Upstage | 41.2% | $0.00053 | $0.0013 | 107 |
| 7 |  Granite 4.2 8BIBM | 36.2% | $0.000077 | $0.00021 | 60 |
| 8 |  Ministral 3 3B 2512Mistral | 26.2% | $0.00019 | $0.00074 | 252 |
| 9 | Granite 4.0 MicroIBM | 25.0% | $0.000051 | $0.0002 | 38 |
| 10 |  Nova Micro 1.0Amazon | 22.5% | $0.000087 | $0.00039 | 180 |
| 11 |  Gemma 4 26B A4BGoogle | 15.0% | $0.0011 | $0.0072 | 172 |
| 12 |  Reka Flash 3rekaai | 1.2% | $0.0008 | $0.067 | 101 |