Calibration

FrontierCode 1.1 Extended → CursorBench 3.2

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

CursorBench 3.2 is estimated from FrontierCode 1.1 Extended with a Michaelis–Menten + offset curve fitted on 5 models measured on both: y = 0.0000 + 2.0000·x / (1.18046 + x), R² = 0.79, cross-validated error 2.1 pp. It is used for 0 estimates.