Calibration

FrontierMath (legacy) → FrontierMath v2 (Tiers 1-3)

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

FrontierMath v2 (Tiers 1-3) is estimated from FrontierMath (legacy) with a linear curve fitted on 6 models measured on both: y = 1.0129·x + -0.0111, R² = 1.00, cross-validated error 0.7 pp. It is used for 1 estimate.

Estimated model	FrontierMath (legacy)	FrontierMath v2 (Tiers 1-3)	Source
Claude Opus 4.7 (Adaptive)	43.8%	43.3%	estimated ± 0.7 pp, low confidence