| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| MMLU-Redux2 | UCPO | PAQ91.67 | 14 | 1mo ago | |
| GPQA Diamond | UCPO | PAQ Score0.677 | 14 | 1mo ago | |
| MMLU-Pro | Accuracy82.93 | 9 | 1mo ago | ||
| LiveBench 2024-11-25 | Accuracy75.9 | 5 | 12d ago | ||
| LiveBench 0831 | Qwen2.5-VL-72B | Accuracy0.57 | 5 | 1mo ago | |
| MMLU-redux | Qwen2.5-72B | Accuracy86.8 | 5 | 1mo ago |