Calibration

FrontierSWE v2 → Vals SWE-bench

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

Vals SWE-bench is estimated from FrontierSWE v2 with a Michaelis–Menten + offset curve fitted on 12 models measured on both: y = 0.7267 + 0.4017·x / (0.30609 + x), R² = 0.63, cross-validated error 4.9 pp. It is used for 1 estimate.

Estimated model	FrontierSWE v2	Vals SWE-bench	Source
Grok 4.7	29.5%	92.4%	estimated ± 4.9 pp, high confidence