| Benchmark | Model | Rank | Results |
|---|---|---|---|
| mathematical-reasoning-on-aime-2025 | Granite-3.3 (8B) PRISM | – | Accuracy: 27.96 |
| mathematical-reasoning-on-aime24 | Granite-3.3 (8B) PRISM | – | Acc: 37.18 |
| mathematical-reasoning-on-math-500 | Granite-4-H Micro (3B) PRISM | – | Accuracy: 82.73 |
| multiple-choice-question-answering-mcqa-on-gpqa-diamond | Granite-3.3 (8B) PRISM (MCS) + RL (MC) | – | Accuracy: 52.86 |