Calibration
ERQA → AA-MMMU-Pro
AA-MMMU-Pro is estimated from ERQA with a linear curve fitted on 8 models measured on both: y = 0.4439·x + 0.4920, R² = 0.70, cross-validated error 2.2 pp. It is used for 3 estimates.
| Estimated model | ERQA | AA-MMMU-Pro | Source |
|---|---|---|---|
| Grok 4.20 | 54.1% | 73.2% | estimated ± 2.2 pp, high confidence |
| Seed 2.1 Pro | 72.0% | 81.2% | estimated ± 2.2 pp, high confidence |
| Seed 2.1 Turbo | 71.3% | 80.9% | estimated ± 2.2 pp, high confidence |