benchgap
Calibration

PostTrainBench v1.1 → FrontierCode 1.1 Main

FrontierCode 1.1 Main is estimated from PostTrainBench v1.1 with a offset logistic curve fitted on 6 models measured on both: y = 0.4075 + (0.5383 − 0.4075) / (1 + exp(−176.43·(x − 0.3304))), R² = 0.95, cross-validated error 2.8 pp. It is used for 3 estimates.

Estimated modelPostTrainBench v1.1FrontierCode 1.1 MainSource
Gemini 3.1 Pro22.0%40.7%estimated ± 2.8 pp, low confidence
Gemini 4 Argon45.3%53.8%estimated ± 2.8 pp, medium confidence
GPT-5.419.0%40.7%estimated ± 2.8 pp, low confidence