Calibration
GPQA Diamond → Artificial Analysis Intelligence Index
Artificial Analysis Intelligence Index is estimated from GPQA Diamond with a inverse Michaelis–Menten curve fitted on 43 models measured on both: y = 0.12974·(x − 0.4132) / (0.4132 + 0.6783 − x), R² = 0.73, cross-validated error 7.4 pp. It is used for 2 estimates.
| Estimated model | GPQA Diamond | Artificial Analysis Intelligence Index | Source |
|---|---|---|---|
| LLaDA2.2-mini | 44.4% | 0.6% | estimated ± 7.4 pp, low confidence |
| Pareto 26.10 Preview | 92.4% | 39.6% | estimated ± 7.4 pp, medium confidence |