| # | Model | Provider | Type | Score | |
|---|---|---|---|---|---|
| 1 | o1 | OpenAI | Proprietary | 91.8% | |
| 2 | GPT-4.1 | OpenAI | Proprietary | 90.2% | |
| 5 | GPT-4.1 mini | OpenAI · Closed | 87.5% | ||
| 6 | Trinity-Large-Preview | Arcee AI · Open weight | 87.2% | ||
| 7 | o3-mini | OpenAI · Closed | 86.9% | ||
| 8 | LongCat-Flash-Lite-Sparse | Meituan · Open weight | 85.3% | ||
| 9 | GPT-4.1 nano | OpenAI · Closed | 80.1% |