xAI · model
Grok 3 [Beta] benchmark scores
As of 2026-11-22, Grok 3 [Beta] (xAI) has measured scores on 5 benchmarks and estimated scores on 22 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 73.4% | estimated ± 6.2 pp, medium confidence |
| AA-HLE | 12.0% | estimated ± 5.7 pp, medium confidence |
| Artificial Analysis Intelligence Index | 11.1% | estimated ± 7.4 pp, medium confidence |
| CritPt | 0.4% | estimated ± 4.0 pp, high confidence |
| GPQA Diamond | 75.6% | estimated ± 1.2 pp, high confidence |
| HLE | 12.6% | estimated ± 6.5 pp, medium confidence |
| AA-Omniscience Accuracy | 31.3% | estimated ± 2.7 pp, low confidence |
| SWE-bench Verified | 56.3% | estimated ± 9.3 pp, low confidence |
| MMLU-Pro | 79.9% | measured |
| AIME 2024 | 52.2% | measured |
| BBH | 95.9% | estimated ± 2.8 pp, medium confidence |
| GPQA | 75.4% | measured |
| MMLU | 88.6% | estimated ± 2.1 pp, high confidence |
| Vals GPQA Diamond | 73.3% | estimated ± 4.5 pp, high confidence |
| Vals MMLU-Pro | 81.2% | estimated ± 2.3 pp, high confidence |
| MATH-500 | 93.5% | estimated ± 1.3 pp, high confidence |
| MMMLU | 85.2% | estimated ± 1.7 pp, high confidence |
| AIME 2025 | 37.1% | estimated ± 3.5 pp, high confidence |
| AA MMLU-Pro | 85.7% | estimated ± 1.2 pp, low confidence |
| ARC-AGI-1 | 36.9% | estimated ± 6.7 pp, low confidence |
| ARC-AGI-2 | 0.0% | estimated ± 10.9 pp, low confidence |
| HLE w/o tools | 12.0% | estimated ± 5.3 pp, medium confidence |
| C-Eval | 89.8% | estimated ± 1.2 pp, high confidence |
| MMLU-Redux | 90.3% | estimated ± 0.6 pp, high confidence |
| SuperGPQA | 55.4% | estimated ± 8.5 pp, low confidence |
| SimpleQA | 43.6% | measured |
| LiveCodeBench | 57.0% | measured |