| Benchmark | Model | Rank | Results |
|---|---|---|---|
| code-generation-on-scicode | Dory1.0-4B | – | Accuracy: 35.11 |
| code-generation-on-swe-bench-verified | Dory1.0-4B (NeMo Gym / Nemo Evaluator SDK) | – | Accuracy: 61.8 |
| code-generation-on-terminal-bench-2-1 | Dory1.0-4B (Terminus-2) | – | Accuracy: 22.2 |
| instruction-following-on-ifbench | Dory1.0-4B | – | Accuracy: 79.0 |
| multiple-choice-question-answering-mcqa-on-gpqa-diamond | Dory1.0-4B | – | Accuracy: 72.41 |