Calibration

Video-MME (with subtitle) → DynaMath

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

DynaMath is estimated from Video-MME (with subtitle) with a Michaelis–Menten curve fitted on 6 models measured on both: y = 1.8669·x / (1.03296 + x), R² = 0.79, cross-validated error 2.8 pp. It is used for 0 estimates.