| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| IFEval, MultiIF, MMLU, MMLU-Pro, GPQA-Diamond, HumanEval, MBPP, GSM8K, MATH-500, TheoremQA, CMMLU, CEval | CHORD | Average Score69.3 | 20 | 1mo ago | |
| Aggregate HumanEval, MBPP, GSM8K, Minerva Math, MMLU, ARC, HellaSwag, PIQA, WinoGrande | Nemotron-Labs-Diffusion-8B | Average Accuracy72.36 | 9 | 18d ago |