benchgap
Calibration

NL2Repo → LiveCodeBench v6

LiveCodeBench v6 is estimated from NL2Repo with a Michaelis–Menten curve fitted on 5 models measured on both: y = 1.3254·x / (0.21727 + x), R² = 0.48, cross-validated error 3.8 pp. It is used for 2 estimates.

Estimated modelNL2RepoLiveCodeBench v6Source
Seed 2.1 Pro47.0%90.6%estimated ± 3.8 pp, low confidence
Seed 2.1 Turbo43.7%88.5%estimated ± 3.8 pp, low confidence