Calibration
GDP.pdf → AA EnterpriseOps-Gym
AA EnterpriseOps-Gym is estimated from GDP.pdf with a linear curve fitted on 9 models measured on both: y = 1.0434·x + 0.2414, R² = 0.62, cross-validated error 4.4 pp. It is used for 10 estimates.
| Estimated model | GDP.pdf | AA EnterpriseOps-Gym | Source |
|---|---|---|---|
| Claude Haiku 5.5 | 20.8% | 45.8% | estimated ± 4.4 pp, high confidence |
| DeepSeek V4.1 Flash | 12.8% | 37.5% | estimated ± 4.4 pp, high confidence |
| Gemini 3.8 Flash | 21.0% | 46.1% | estimated ± 4.4 pp, high confidence |
| Gemini 4 Argon | 21.8% | 46.9% | estimated ± 4.4 pp, high confidence |
| GPT-6 Luna | 22.8% | 47.9% | estimated ± 4.4 pp, medium confidence |
| GPT-6 Sol | 25.2% | 50.4% | estimated ± 4.4 pp, medium confidence |
| Grok 4.7 | 20.0% | 45.0% | estimated ± 4.4 pp, high confidence |
| MiMo-V2.6-Pro | 19.2% | 44.2% | estimated ± 4.4 pp, high confidence |
| Mistral Large 4 | 18.6% | 43.6% | estimated ± 4.4 pp, high confidence |
| Muse Spark 1.3 | 26.6% | 51.9% | estimated ± 4.4 pp, medium confidence |