Calibration
Vals GPQA Diamond → Artificial Analysis Intelligence Index
Artificial Analysis Intelligence Index is estimated from Vals GPQA Diamond with a offset logistic curve fitted on 47 models measured on both: y = 0.1683 + (0.5117 − 0.1683) / (1 + exp(−27.20·(x − 0.8895))), R² = 0.60, cross-validated error 7.9 pp. It is used for 7 estimates.
| Estimated model | Vals GPQA Diamond | Artificial Analysis Intelligence Index | Source |
|---|---|---|---|
| Claude Haiku 4.5 | 72.2% | 17.2% | estimated ± 7.9 pp, medium confidence |
| Gemini 3.1 Flash-Lite | 81.1% | 20.5% | estimated ± 7.9 pp, medium confidence |
| GLM-4.5 | 72.2% | 17.2% | estimated ± 7.9 pp, medium confidence |
| Laguna M.1 | 27.0% | 16.8% | estimated ± 7.9 pp, low confidence |
| Laguna XS.2 | 55.1% | 16.8% | estimated ± 7.9 pp, medium confidence |
| MiMo-V2.5 | 81.6% | 20.9% | estimated ± 7.9 pp, medium confidence |
| Qwen3.5 Flash | 82.8% | 22.3% | estimated ± 7.9 pp, medium confidence |