Calibration

MMLU-Pro → C-Eval

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

C-Eval is estimated from MMLU-Pro with a Michaelis–Menten curve fitted on 6 models measured on both: y = 2.0000·x / (1.03139 + x), R² = 0.90, cross-validated error 1.0 pp. It is used for 1 estimate.

Estimated model	MMLU-Pro	C-Eval	Source
Solar Open 2	86.2%	91.1%	estimated ± 1.0 pp, medium confidence