Calibration

LiveCodeBench → SWE-bench Verified

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

SWE-bench Verified is estimated from LiveCodeBench with a Michaelis–Menten curve fitted on 7 models measured on both: y = 1.9265·x / (1.33957 + x), R² = 0.95, cross-validated error 3.5 pp. It is used for 0 estimates.