Calibration
Vals GPQA Diamond → ARC-AGI-1
ARC-AGI-1 is estimated from Vals GPQA Diamond with a Hill curve fitted on 21 models measured on both: y = 0.0000 + (1.2000 − 0.0000)·x^6.00 / (0.76013^6.00 + x^6.00), R² = 0.76, cross-validated error 5.9 pp. It is used for 7 estimates.
| Estimated model | Vals GPQA Diamond | ARC-AGI-1 | Source |
|---|---|---|---|
| Claude Haiku 4.5 | 72.2% | 50.8% | estimated ± 5.9 pp, low confidence |
| Gemini 3.1 Flash-Lite | 81.1% | 71.5% | estimated ± 5.9 pp, medium confidence |
| GLM-4.5 | 72.2% | 50.8% | estimated ± 5.9 pp, low confidence |
| Laguna M.1 | 27.0% | 0.2% | estimated ± 5.9 pp, low confidence |
| Laguna XS.2 | 55.1% | 15.2% | estimated ± 5.9 pp, low confidence |
| MiMo-V2.5 | 81.6% | 72.6% | estimated ± 5.9 pp, medium confidence |
| Qwen3.5 Flash | 82.8% | 75.1% | estimated ± 5.9 pp, medium confidence |