| Rank | Compare | Model | Score |
|---|---|---|---|
| 51 | 30.7% | ||
| 52 |  Ministral 3 14B Mistral | 30.0% | |
| 53 | 26.7% | ||
| 54 |  Exaone 4.0 1.2B (Non-reasoning) LG AI Research | 24.0% | |
| 55 | Ministral 3 3B Mistral | 22.0% | |
| 56 |  Llama 4 Maverick Meta | 19.3% | |
| 57 |  Phi-4 Microsoft | 18.0% | |
| 58 |  Nova Premier Amazon | 17.3% | |
| 59 |  Hermes 4 - Llama-3.1 405B (Non-reasoning) Nous Research | 15.3% | |
| 60 | Llama 4 Scout Meta | 14.0% | |
| 61 | 13.7% | ||
| 62 | 13.3% | ||
| 63 |  Command A Cohere | 13.0% | |
| 64 | Hermes 4 - Llama-3.1 70B (Non-reasoning) Nous Research | 11.3% | |
| 65 |  Jamba Reasoning 3B AI21 Labs | 10.7% | |
| 66 | 7.7% | ||
| 67 | Phi-4 Mini Instruct Microsoft | 6.7% | |
| 68 | 6.3% | ||
| 69 | 6.0% | ||
| 70 | Nova Micro Amazon | 6.0% | |
| 71 | 3.0% | ||
| 72 |  Gemma 3 270M Google | 2.3% | |
| 73 | Jamba 1.7 Large AI21 Labs | 2.3% | |
| 74 | 1.7% | ||
| 75 | 1.3% | ||
| 76 | Jamba 1.7 Mini AI21 Labs | 0.3% | |
| 77 | 0.0% | ||
| 78 |  Molmo 7B-D Allen Institute for AI | 0.0% |
