aime2025 Arcee
6 models evaluated
|
| 1 | Claude Opus 4.6 | Anthropic | Closed | 99.8% | |
| 2 | Kimi K2.5 | Moonshot AI | Open | 96.3% | |
| 3 | Trinity-Large-Thinking | Arcee AI | Open | 96.3% | |
| 4 | GLM-5 | Z.AI | Open | 93.3% | |
| 5 | MiniMax M2.7 | MiniMax | Open | 80.0% | |
| 6 | Trinity-Large-Preview | Arcee AI | Open | 24.0% | |