xAI · model
Grok 4 benchmark scores
As of 2026-10-07, Grok 4 (xAI) has measured scores on 13 benchmarks and estimated scores on 66 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 87.7% | measured |
| AA-HLE | 26.7% | measured |
| AA-SciCode | 46.0% | estimated ± 8.6 pp, low confidence |
| AIME26 | 95.5% | estimated ± 0.6 pp, low confidence |
| Artificial Analysis Intelligence Index | 22.5% | measured |
| CritPt | 2.0% | measured |
| GDPval-AA | 17.9% | estimated ± 8.8 pp, medium confidence |
| GPQA Diamond | 85.2% | estimated ± 2.2 pp, high confidence |
| HLE | 30.9% | estimated ± 9.5 pp, medium confidence |
| IFBench | 58.5% | estimated ± 6.9 pp, medium confidence |
| AA-LCR | 68.0% | measured |
| AA-Omniscience Accuracy | 40.5% | measured |
| BrowseComp | 59.2% | estimated ± 5.7 pp, medium confidence |
| IFEval | 92.0% | estimated ± 1.3 pp, high confidence |
| LongBench v2 | 61.5% | estimated ± 5.9 pp, medium confidence |
| τ²-bench results | 74.9% | measured |
| AA Coding Index | 41.6% | estimated ± 11.4 pp, low confidence |
| AA-MMMU-Pro | 68.8% | measured |
| APEX-Agents-AA | 1.6% | estimated ± 11.4 pp, low confidence |
| BioMysteryBench (human-difficult) | 41.2% | estimated ± 5.9 pp, low confidence |
| BFCL v4 | 53.2% | estimated ± 9.8 pp, low confidence |
| CyberGym | 35.1% | estimated ± 10.0 pp, low confidence |
| τ³-bench results | 69.0% | estimated ± 8.9 pp, medium confidence |
| WideResearch | 63.4% | estimated ± 5.9 pp, low confidence |
| AA Agentic Index | 11.9% | estimated ± 6.4 pp, medium confidence |
| MMLU-Pro | 85.7% | estimated ± 3.5 pp, high confidence |
| FrontierMath v2 (Tier 4) | 2.1% | measured |
| FrontierMath v2 (Tiers 1-3) | 19.7% | measured |
| GPQA | 87.1% | estimated ± 3.2 pp, high confidence |
| AA-IFBench | 53.7% | measured |
| Gert Labs | 42.3% | measured |
| AA EnterpriseOps-Gym | 16.5% | estimated ± 3.6 pp, low confidence |
| ApprenticeBench | 0.0% | estimated ± 3.1 pp, low confidence |
| ARC-AGI-1 | 63.8% | estimated ± 5.4 pp, medium confidence |
| ARC-AGI-2 | 24.9% | estimated ± 14.2 pp, low confidence |
| OfficeQA Pro | 50.3% | estimated ± 7.2 pp, low confidence |
| OSWorld-Verified | 55.4% | estimated ± 5.9 pp, medium confidence |
| Vals GPQA Diamond | 84.7% | estimated ± 7.4 pp, medium confidence |
| Vals LiveCodeBench | 81.5% | estimated ± 3.8 pp, low confidence |
| Vals MMLU-Pro | 86.9% | estimated ± 2.2 pp, high confidence |
| Vals SWE-bench | 74.6% | estimated ± 1.4 pp, medium confidence |
| AA ITBench | 64.6% | estimated ± 3.6 pp, low confidence |
| HLE w/o tools | 23.6% | estimated ± 5.7 pp, medium confidence |
| OSWorld 2.0 | 0.0% | estimated ± 4.5 pp, low confidence |
| Vibe Code Bench | 27.7% | estimated ± 6.5 pp, low confidence |
| CharXiv | 71.0% | estimated ± 7.5 pp, low confidence |
| CharXiv w/o tools | 75.5% | estimated ± 2.4 pp, medium confidence |
| C-Eval | 92.1% | estimated ± 1.0 pp, medium confidence |
| Claw-Eval | 56.5% | estimated ± 6.8 pp, low confidence |
| HMMT Nov 2025 | 96.8% | estimated ± 2.6 pp, low confidence |
| MathVision | 81.5% | estimated ± 4.9 pp, high confidence |
| MCP Atlas | 43.5% | estimated ± 14.8 pp, low confidence |
| MCP-Tasks | 79.1% | estimated ± 8.2 pp, low confidence |
| MMAnswerBench | 82.4% | estimated ± 1.3 pp, low confidence |
| MMLU-Redux | 96.7% | estimated ± 1.1 pp, medium confidence |
| MMMU-Pro | 74.1% | estimated ± 4.4 pp, medium confidence |
| ScreenSpot Pro | 67.8% | estimated ± 10.7 pp, low confidence |
| SuperGPQA | 81.4% | estimated ± 7.5 pp, medium confidence |
| Toolathlon | 36.6% | estimated ± 7.2 pp, medium confidence |
| ERQA | 21.9% | estimated ± 3.9 pp, medium confidence |
| HealthBench Hard | 12.5% | estimated ± 7.7 pp, low confidence |
| MedXpertQA (MM) | 42.5% | estimated ± 6.5 pp, low confidence |
| MMLU-Pro (Arcee) | 84.6% | estimated ± 4.0 pp, medium confidence |
| React Native Evals | 72.6% | measured |
| ResearchClawBench | 14.7% | estimated ± 2.2 pp, medium confidence |
| ARC-AGI-3 | 1.7% | estimated ± 3.4 pp, medium confidence |
| FrontierMath (legacy) | 20.5% | estimated ± 0.7 pp, low confidence |
| BioMysteryBench (human-solvable) | 81.7% | estimated ± 0.6 pp, low confidence |
| HealthBench (raw) | 35.7% | estimated ± 6.4 pp, low confidence |
| HealthBench (length-adjusted) | 53.9% | estimated ± 3.7 pp, low confidence |
| HealthBench Professional | 57.7% | estimated ± 4.8 pp, low confidence |
| HealthBench Professional (raw) | 47.0% | estimated ± 5.0 pp, low confidence |
| HLE-Verified | 15.3% | estimated ± 2.0 pp, low confidence |
| LABBench2 | 73.5% | estimated ± 1.8 pp, low confidence |
| MMMU | 78.3% | estimated ± 0.9 pp, high confidence |
| MMLU | 93.5% | estimated ± 1.1 pp, low confidence |
| MMMU-Pro w/ Python | 74.6% | estimated ± 1.6 pp, high confidence |
| RealWorldQA | 81.4% | estimated ± 1.2 pp, low confidence |
| OmniDocBench 1.5 | 0.0% | estimated ± 6.4 pp, low confidence |