benchgap
Calibration

QwenClawBench → τ²-bench results

τ²-bench results is estimated from QwenClawBench with a offset logistic curve fitted on 10 models measured on both: y = 0.7006 + (0.9598 − 0.7006) / (1 + exp(−200.00·(x − 0.5177))), R² = 0.79, cross-validated error 3.6 pp. It is used for 0 estimates.