Calibration
EnterpriseOps-Gym → τ³-Bench Banking
τ³-Bench Banking is estimated from EnterpriseOps-Gym with a Hill curve fitted on 10 models measured on both: y = 0.0000 + (0.5396 − 0.0000)·x^6.00 / (0.32414^6.00 + x^6.00), R² = 0.58, cross-validated error 11.4 pp. It is used for 3 estimates.
| Estimated model | EnterpriseOps-Gym | τ³-Bench Banking | Source |
|---|---|---|---|
| Gemini 3.7 Flash (medium) | 50.4% | 50.4% | estimated ± 11.4 pp, low confidence |
| Gemini 3.5 Flash (high) | 50.1% | 50.3% | estimated ± 11.4 pp, low confidence |
| DeepSeek V4 Pro 0813 (max) | 49.6% | 50.1% | estimated ± 11.4 pp, low confidence |