Calibration
WideResearch → τ³-bench results
τ³-bench results is estimated from WideResearch with a offset logistic curve fitted on 7 models measured on both: y = 0.6797 + (0.0000 − 0.6797) / (1 + exp(−200.00·(x − 0.8212))), R² = 0.96, cross-validated error 11.4 pp. It is used for 2 estimates.
| Estimated model | WideResearch | τ³-bench results | Source |
|---|---|---|---|
| dots3-note Preview | 78.9% | 67.9% | estimated ± 11.4 pp, low confidence |
| Qwen3.8 Max | 81.9% | 41.2% | estimated ± 11.4 pp, low confidence |