| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| GPQA | Accuracy85.7 | 49 | 1mo ago | ||
| MMLU-Pro | Accuracy90.1 | 48 | 23d ago | ||
| GPQA D | Length Scaling | Accuracy69.9 | 33 | 3mo ago | |
| SciBench | SC | Accuracy (SciBench)28.3 | 24 | 1mo ago | |
| GPQA Diamond | VP2O | Accuracy67 | 19 | 1mo ago | |
| MMLU-Pro (test) | Base | Accuracy41.9 | 18 | 2mo ago | |
| ARC-Challenge, ARC-Easy, OpenBookQA | MobileMoE-L | ARC-C Accuracy57.9 | 13 | 1mo ago | |
| GPQA | CapFlow | Solve Rate42.41 | 11 | 4mo ago | |
| ARC-E | MobileMoE-L | Accuracy81.7 | 7 | 1mo ago | |
| SuperGPQA | Score73.8 | 5 | 23d ago | ||
| HealthBench Hard | Score42 | 5 | 23d ago | ||
| OBQA | MobileLLM-Pro | Accuracy42.8 | 4 | 1mo ago | |
| ARC-C | MobileMoE-L | Accuracy55.1 | 4 | 1mo ago | |
| GPQA D | AIPO | Avg@472.55 | 4 | 2mo ago | |
| GPQA Diamond | DeepSeek-R1 | Pass@1 Score71.5 | 4 | 2mo ago | |
| ChemBench | AgentSPEX | Score83.3 | 3 | 3mo ago | |
| StemEZ physical chemistry MMLU-Pro | AgentSPEX | Score86.57 | 3 | 3mo ago | |
| SciBench chemistry | AgentSPEX | Score90.61 | 3 | 3mo ago |
