Calibration
APEX-Agents-AA → τ²-bench results
τ²-bench results is estimated from APEX-Agents-AA with a Michaelis–Menten curve fitted on 22 models measured on both: y = 0.9676·x / (0.00511 + x), R² = 0.68, cross-validated error 7.3 pp. It is used for 4 estimates.
| Estimated model | APEX-Agents-AA | τ²-bench results | Source |
|---|---|---|---|
| Apodex 1.1 | 31.2% | 95.2% | estimated ± 7.3 pp, medium confidence |
| Apodex 1.1 Mini | 31.2% | 95.2% | estimated ± 7.3 pp, medium confidence |
| GPT-5.6 Luna | 35.8% | 95.4% | estimated ± 7.3 pp, medium confidence |
| Qwen3.8 Max Preview | 42.4% | 95.6% | estimated ± 7.3 pp, medium confidence |