Calibration
ExploitBench Internal Port → ExploitBench
ExploitBench is estimated from ExploitBench Internal Port with a Michaelis–Menten curve fitted on 7 models measured on both: y = 1.0591·x / (0.02302 + x), R² = 0.97, cross-validated error 7.0 pp. It is used for 0 estimates.
Fitted on
| Model | ExploitBench Internal Port | ExploitBench |
|---|---|---|
| GPT-5.6 Sol | 5.5% | 73.5% |
| GPT-6.1 Sol | 21.5% | 99.7% |
| GPT-6.1 Terra | 24.5% | 97.1% |
| GPT-6.2 Astra | 53.5% | 100.0% |
| GPT-6.2 Sol | 50.5% | 100.0% |
| GPT-6 Astra | 39.0% | 100.0% |
| GPT-6 Cyber | 42.0% | 100.0% |