Calibration
Toolathlon-Verified → GDPval-AA
GDPval-AA is estimated from Toolathlon-Verified with a inverse Michaelis–Menten curve fitted on 13 models measured on both: y = 0.89811·(x − 0.1615) / (0.1615 + 1.5070 − x), R² = 0.87, cross-validated error 4.6 pp. It is used for 2 estimates.
| Estimated model | Toolathlon-Verified | GDPval-AA | Source |
|---|---|---|---|
| GLM-5.3-Flash | 78.4% | 63.2% | estimated ± 4.6 pp, high confidence |
| Laguna S 2.1 | 49.7% | 25.7% | estimated ± 4.6 pp, medium confidence |