| Benchmark | Model | Rank | Results |
|---|---|---|---|
| mathematical-reasoning-on-omni-math | Confident Decoding (Qwen3.5-122B-A10B) | #1 | Accuracy: 78.5 |
| question-answering-on-humanity-s-last-exam-hle | Confident Decoding (Qwen3.5-27B) | #34 | Accuracy: 16.8 |
| multiple-choice-question-answering-mcqa-on-gpqa-diamond | Qwen3.5-35B-A3B + Confident Decoding | #48 | Accuracy: 82.8 |