Calibration
CyberGym → Toolathlon-Verified
Toolathlon-Verified is estimated from CyberGym with a Michaelis–Menten curve fitted on 7 models measured on both: y = 0.9350·x / (0.22751 + x), R² = 0.43, cross-validated error 2.2 pp. It is used for 9 estimates.
| Estimated model | CyberGym | Toolathlon-Verified | Source |
|---|---|---|---|
| Claude Mythos 5 | 83.8% | 73.5% | estimated ± 2.2 pp, low confidence |
| Claude Mythos Preview | 83.1% | 73.4% | estimated ± 2.2 pp, low confidence |
| Gemini 3.5 Flash Cyber | 83.2% | 73.4% | estimated ± 2.2 pp, low confidence |
| Gemini 3.8 Flash Cyber | 86.2% | 74.0% | estimated ± 2.2 pp, low confidence |
| GLM-5 | 43.2% | 61.2% | estimated ± 2.2 pp, low confidence |
| GLM-5.1 | 68.7% | 70.2% | estimated ± 2.2 pp, low confidence |
| Ling 3.1 Flash | 87.9% | 74.3% | estimated ± 2.2 pp, low confidence |
| Muse Spark | 43.5% | 61.4% | estimated ± 2.2 pp, low confidence |
| Fugu Cyber | 86.9% | 74.1% | estimated ± 2.2 pp, low confidence |