Calibration

AA EnterpriseOps-Gym → Toolathlon-Verified

Name: benchgap: LLM benchmark scores, measured and estimated
Creator: benchgap

Toolathlon-Verified is estimated from AA EnterpriseOps-Gym with a offset logistic curve fitted on 5 models measured on both: y = 1.0000 + (0.7360 − 1.0000) / (1 + exp(−200.00·(x − 0.3245))), R² = 0.95, cross-validated error 2.6 pp. It is used for 0 estimates.