| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| GPQA D | Length Scaling | Accuracy69.9 | 33 | 16d ago | |
| GPQA | Accuracy85.7 | 25 | 1mo ago | ||
| GPQA | CapFlow | Solve Rate42.41 | 11 | 1mo ago | |
| ChemBench | AgentSPEX | Score83.3 | 3 | 3d ago | |
| StemEZ physical chemistry MMLU-Pro | AgentSPEX | Score86.57 | 3 | 3d ago | |
| SciBench chemistry | AgentSPEX | Score90.61 | 3 | 3d ago |
