| Benchmark | Model | Rank | Results |
|---|---|---|---|
| code-generation-on-livecodebench-v6 | TrOPD (Qwen3-SFT-1.7B) | – | Pass@1: 36.0 |
| instruction-following-on-ifbench | TrOPD (Qwen3-SFT-1.7B) | – | Accuracy: 42.18 |
| mathematical-reasoning-on-aime-2025 | TrOPD (Qwen3-SFT-1.7B) | – | Accuracy: 44.06 |
| mathematical-reasoning-on-aime24 | TrOPD (Qwen3-SFT-1.7B) | – | Acc: 52.08 |
| multiple-choice-question-answering-mcqa-on-gpqa-diamond | TrOPD (DeepSeek-R1-Distill-Qwen-1.5B) | – | Accuracy: 36.24 |