# MMMLU → CritPt

Calibration · https://benchgap.net/calibration/mmmlu/current/to/critpt/current · data retrieved 2026-11-18

CritPt is estimated from MMMLU with a linear curve fitted on 6 models measured on both: y = 1.8150·x + -1.5283, R² = 0.86, cross-validated error 2.9 pp. It is used for 3 estimates.

| Estimated model | MMMLU | CritPt | Source |
|---|---|---|---|
| [Interfaze Beta](https://benchgap.net/model/interfaze-beta.md) | 90.9% | 12.2% | estimated ± 2.9 pp, low confidence |
| [Claude 4 Opus](https://benchgap.net/model/claude-4-opus.md) | 87.4% | 5.8% | estimated ± 2.9 pp, medium confidence |
| [Qwen3 32B](https://benchgap.net/model/qwen3-32b.md) | 76.5% | 0.0% | estimated ± 2.9 pp, low confidence |

## Fitted on

| Model | MMMLU | CritPt |
|---|---|---|
| [Claude 3.7 Sonnet](https://benchgap.net/model/claude-3-7-sonnet.md) | 83.2% | 0.0% |
| [Claude 3.7 Sonnet Thinking](https://benchgap.net/model/claude-3-7-sonnet-thinking.md) | 86.1% | 0.9% |
| [Gemma 4 12B](https://benchgap.net/model/gemma-4-12b.md) | 83.4% | 0.0% |
| [K-EXAONE 2.0](https://benchgap.net/model/k-exaone-2-0.md) | 86.6% | 0.9% |
| [Qwen3.7 Max](https://benchgap.net/model/qwen3-7-max.md) | 90.3% | 13.4% |
| [Qwen3.7 Plus](https://benchgap.net/model/qwen3-7-plus.md) | 89.0% | 9.1% |
