| Rang | Vergleichen | Modell | Ergebnis |
|---|---|---|---|
| 1 |  Gemini 3.1 Pro Preview Google | 93.2% | |
| 2 | 90.0% | ||
| 3 | 89.1% | ||
| 4 | 87.6% | ||
| 5 | 86.5% | ||
| 6 | 86.5% | ||
| 7 | 85.9% | ||
| 8 |  gpt-oss-120b (High) OpenAI | 82.8% | |
| 9 |  Llama 4 Maverick Meta | 82.5% | |
| 10 | 81.9% | ||
| 11 |  Solar Open 100B (Reasoning) Upstage | 81.6% | |
| 12 |  Magistral Medium 1.2 Mistral | 81.6% | |
| 13 |  Nova 2.0 Omni (Low) Amazon | 81.0% | |
| 14 | Nova 2.0 Omni (Medium) Amazon | 81.0% | |
| 15 |  INTELLECT-3 (Based on GLM-4.5-Air) Prime Intellect | 80.3% | |
| 16 | gpt-oss-120b (Low) OpenAI | 80.1% | |
| 17 |  Doubao Seed Code ByteDance Seed | 79.8% | |
| 18 | Nova Premier Amazon | 79.6% | |
| 19 |  Qwen3 Coder Next Alibaba | 79.4% | |
| 20 | 78.6% | ||
| 21 | 78.6% | ||
| 22 |  K2-V2 (High) Institute of Foundation Models | 78.6% | |
| 23 | 78.2% | ||
| 24 | K2-V2 (Medium) Institute of Foundation Models | 76.7% | |
| 25 | 76.4% | ||
| 26 | Magistral Small 1.2 Mistral | 76.3% | |
| 27 | gpt-oss-20b (High) OpenAI | 75.2% | |
| 28 |  Mi:dm K 2.5 Pro Korea Telecom | 74.2% | |
| 29 | Llama 4 Scout Meta | 74.1% | |
| 30 |  EXAONE 4.0 32B (Reasoning) LG AI Research | 73.5% | |
| 31 | gpt-oss-20b (Low) OpenAI | 72.8% | |
| 32 | 72.7% | ||
| 33 |  Olmo 3.1 32B Think Allen Institute for AI | 71.5% | |
| 34 | K2-V2 (Low) Institute of Foundation Models | 71.4% | |
| 35 | Ministral 3 14B Mistral | 68.3% | |
| 36 | Ministral 3 8B Mistral | 66.3% | |
| 37 | 61.2% | ||
| 38 | Ministral 3 3B Mistral | 56.7% | |
| 39 |  Tiny Aya Global Cohere | 55.1% |
