xAI · model
Grok 2 benchmark scores
As of 2026-10-29, Grok 2 (xAI) has measured scores on 4 benchmarks and estimated scores on 31 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 53.9% | estimated ± 6.3 pp, medium confidence |
| AA-HLE | 4.7% | estimated ± 5.0 pp, high confidence |
| CritPt | 0.0% | estimated ± 3.4 pp, medium confidence |
| GPQA Diamond | 56.4% | estimated ± 1.2 pp, high confidence |
| HLE | 3.6% | estimated ± 6.9 pp, low confidence |
| AA-MMMU-Pro | 50.6% | estimated ± 3.0 pp, high confidence |
| MMLU-Pro | 75.5% | measured |
| Artificial Analysis Intelligence Index | 11.5% | estimated ± 2.1 pp, high confidence |
| BBH | 92.7% | estimated ± 2.4 pp, low confidence |
| GPQA | 56.0% | measured |
| MMLU | 87.5% | measured |
| AA-Omniscience Accuracy | 20.5% | estimated ± 1.5 pp, high confidence |
| MMMLU | 76.1% | estimated ± 1.7 pp, high confidence |
| MMMU | 66.1% | measured |
| AA MMLU-Pro | 82.7% | estimated ± 1.0 pp, low confidence |
| HLE w/o tools | 11.6% | estimated ± 5.2 pp, low confidence |
| AI2D_TEST | 88.1% | estimated ± 2.0 pp, high confidence |
| C-Eval | 84.1% | estimated ± 1.2 pp, high confidence |
| CountBench | 93.9% | estimated ± 3.2 pp, high confidence |
| DynaMath | 80.9% | estimated ± 5.8 pp, low confidence |
| ERQA | 47.1% | estimated ± 5.2 pp, low confidence |
| MathVision | 63.7% | estimated ± 3.6 pp, high confidence |
| MLVU (M-Avg) | 79.9% | estimated ± 2.4 pp, high confidence |
| MMLU-Redux | 87.7% | estimated ± 0.6 pp, high confidence |
| MMMU-Pro | 52.5% | estimated ± 2.3 pp, high confidence |
| RealWorldQA | 77.8% | estimated ± 3.6 pp, high confidence |
| ScreenSpot Pro | 53.0% | estimated ± 7.3 pp, low confidence |
| SimpleVQA | 48.9% | estimated ± 7.1 pp, medium confidence |
| SuperGPQA | 55.7% | estimated ± 12.1 pp, low confidence |
| Video-MME (w/o subtitle) | 68.8% | estimated ± 1.3 pp, high confidence |
| Video-MME (with subtitle) | 75.6% | estimated ± 3.2 pp, high confidence |
| VideoMMMU | 57.7% | estimated ± 3.8 pp, high confidence |
| We-Math | 72.9% | estimated ± 7.2 pp, low confidence |
| SimpleQA | 27.0% | estimated ± 13.1 pp, low confidence |
| MMVU | 64.1% | estimated ± 3.8 pp, medium confidence |