xAI · model
Grok 4.3 benchmark scores
As of 2026-11-21, Grok 4.3 (xAI) has measured scores on 31 benchmarks and estimated scores on 142 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 90.1% | measured |
| AA-HLE | 37.2% | measured |
| AA-SciCode | 48.3% | measured |
| AIME26 | 95.9% | estimated ± 1.2 pp, medium confidence |
| Artificial Analysis Intelligence Index | 37.6% | measured |
| CritPt | 8.0% | measured |
| GDPval-AA | 29.2% | measured |
| GPQA Diamond | 90.1% | estimated ± 0.1 pp, high confidence |
| HLE | 35.0% | measured |
| IFBench | 81.3% | measured |
| AA-LCR | 64.3% | measured |
| LiveCodeBench v6 | 88.5% | estimated ± 2.2 pp, medium confidence |
| AA-Omniscience Accuracy | 34.6% | measured |
| SciCode | 47.3% | measured |
| Terminal-Bench 2.1 | 39.6% | estimated ± 6.7 pp, low confidence |
| BrowseComp | 71.8% | estimated ± 4.2 pp, high confidence |
| HLE w/ tools | 53.1% | estimated ± 4.1 pp, medium confidence |
| IFEval | 93.2% | estimated ± 2.8 pp, medium confidence |
| LongBench v2 | 55.8% | estimated ± 5.5 pp, medium confidence |
| τ²-bench results | 97.7% | measured |
| AA Coding Index | 42.3% | measured |
| AA-MMMU-Pro | 78.1% | measured |
| APEX-Agents | 21.5% | estimated ± 4.9 pp, low confidence |
| APEX-Agents-AA | 17.0% | measured |
| BioMysteryBench (human-difficult) | 41.9% | estimated ± 6.8 pp, low confidence |
| DeepSearchQA | 82.2% | estimated ± 4.5 pp, high confidence |
| SWE-bench Verified | 77.4% | estimated ± 4.4 pp, high confidence |
| BFCL v4 | 74.2% | estimated ± 2.7 pp, high confidence |
| CyberGym | 58.8% | estimated ± 5.3 pp, low confidence |
| JobBench | 16.8% | estimated ± 8.2 pp, low confidence |
| SWE-bench Pro | 57.2% | estimated ± 4.7 pp, high confidence |
| WideResearch | 66.2% | estimated ± 5.9 pp, low confidence |
| AA Agentic Index | 17.2% | measured |
| MMLU-Pro | 84.9% | estimated ± 4.4 pp, high confidence |
| AA-IFBench | 81.3% | measured |
| FailureSensorIQ | 65.5% | estimated ± 1.7 pp, high confidence |
| GPQA | 90.1% | measured |
| MMLU | 90.5% | estimated ± 0.9 pp, medium confidence |
| Vals GPQA Diamond | 91.4% | measured |
| Vals LiveCodeBench | 84.5% | measured |
| Vals MMLU-Pro | 85.8% | measured |
| MMMLU | 89.1% | estimated ± 0.7 pp, medium confidence |
| MMMU | 85.0% | estimated ± 0.8 pp, high confidence |
| Chess Puzzles | 18.0% | measured |
| AA MMLU-Pro | 86.5% | estimated ± 0.7 pp, high confidence |
| Gert Labs | 43.9% | measured |
| AA Harvey LAB v1.0 | 80.1% | estimated ± 1.2 pp, low confidence |
| AA Tau3 Banking | 39.2% | estimated ± 2.7 pp, low confidence |
| AA Terminal-Bench 2.1 | 79.4% | estimated ± 2.5 pp, medium confidence |
| ARC-AGI-1 | 89.0% | estimated ± 6.5 pp, medium confidence |
| ARC-AGI-2 | 53.8% | estimated ± 10.9 pp, low confidence |
| CursorBench 3.1 | 39.8% | estimated ± 3.4 pp, low confidence |
| CursorBench 3.2 | 27.9% | estimated ± 4.4 pp, medium confidence |
| EBR-bench | 5.8% | estimated ± 9.1 pp, low confidence |
| FrontierCode 1.1 Main | 15.2% | estimated ± 4.6 pp, medium confidence |
| FrontierMath v2 (Tier 4) | 14.6% | measured |
| FrontierMath v2 (Tiers 1-3) | 42.8% | measured |
| FrontierSWE v2 | 6.8% | estimated ± 9.5 pp, low confidence |
| Furniture Assembly | 26.4% | estimated ± 14.1 pp, low confidence |
| MaxxBench | 39.1% | estimated ± 10.3 pp, low confidence |
| MazeBench | 0.4% | estimated ± 5.5 pp, low confidence |
| Mystery Game Puzzles | 14.7% | estimated ± 9.8 pp, medium confidence |
| NYT Connections | 55.2% | measured |
| OfficeQA Pro | 50.5% | estimated ± 6.8 pp, medium confidence |
| OSWorld-Verified | 53.6% | estimated ± 5.1 pp, low confidence |
| ParseBench-1 | 62.9% | estimated ± 6.2 pp, medium confidence |
| Terminal-Bench 4.1 | 5.6% | estimated ± 14.9 pp, low confidence |
| Terminal-Bench Hard | 43.9% | estimated ± 3.9 pp, medium confidence |
| Vals SWE-bench | 71.4% | measured |
| Terminal-Bench 2.1 (Vals AI) | 41.9% | measured |
| VulcanBench v3 | 83.3% | estimated ± 5.6 pp, low confidence |
| CWE-Bench-AA | 36.6% | estimated ± 10.1 pp, low confidence |
| AA Cyber Index | 11.8% | estimated ± 9.2 pp, low confidence |
| DeepsecBench-AA | 3.1% | estimated ± 8.9 pp, low confidence |
| HealthBench Professional (length-adjusted, Artificial Analysis) | 57.3% | estimated ± 4.3 pp, medium confidence |
| HealthBench Professional (raw, Artificial Analysis) | 51.1% | estimated ± 0.8 pp, low confidence |
| AA ITBench | 36.7% | estimated ± 2.8 pp, high confidence |
| Bug Hunt Bench | 20.8% | estimated ± 10.5 pp, low confidence |
| Chartography (tools) | 82.2% | estimated ± 4.1 pp, low confidence |
| CursorBench 4.0 | 30.8% | estimated ± 3.4 pp, medium confidence |
| CWE-bench v1 | 38.7% | estimated ± 6.4 pp, medium confidence |
| DeepSWE | 56.3% | estimated ± 7.2 pp, low confidence |
| DRACO | 75.5% | estimated ± 2.0 pp, medium confidence |
| FrontierCode 2.0 | 26.1% | estimated ± 9.3 pp, low confidence |
| FrontierPhysics | 0.0% | estimated ± 6.1 pp, low confidence |
| HLE w/o tools | 32.9% | estimated ± 4.9 pp, high confidence |
| OSWorld 2.0 | 0.0% | estimated ± 5.9 pp, low confidence |
| OSWorld 2.1 | 44.5% | estimated ± 6.0 pp, low confidence |
| OSWorld 2.1 Offline | 5.9% | estimated ± 1.9 pp, low confidence |
| PostTrainBench v1.1 | 16.6% | estimated ± 3.7 pp, medium confidence |
| SpireClimb | 5.8% | estimated ± 7.2 pp, low confidence |
| StealthMark | 35.9% | estimated ± 9.7 pp, low confidence |
| Toolathlon-Verified | 44.0% | estimated ± 2.3 pp, low confidence |
| VulcanBench Frontier v4 | 78.5% | estimated ± 6.8 pp, low confidence |
| GDP.pdf | 17.0% | estimated ± 4.6 pp, medium confidence |
| AA Terminal-Bench 4.0 | 0.0% | estimated ± 13.7 pp, low confidence |
| ARC-AGI-3 | 0.0% | estimated ± 8.3 pp, low confidence |
| ARC-AGI-3 Provider Adapter | 2.7% | estimated ± 7.3 pp, low confidence |
| cursorBench41 | 44.8% | estimated ± 3.7 pp, medium confidence |
| MKA | 24.3% | estimated ± 6.4 pp, low confidence |
| SkillsBench official leaderboard (BenchFlow) | 41.7% | measured |
| BinaryAudit | 25.9% | estimated ± 12.0 pp, low confidence |
| OTelBench | 22.5% | estimated ± 6.6 pp, medium confidence |
| Vibe Code Bench | 16.1% | estimated ± 10.0 pp, low confidence |
| FACTS Multimodal | 32.1% | estimated ± 4.0 pp, medium confidence |
| Hermes Bench | 59.3% | estimated ± 2.9 pp, high confidence |
| Hermes Index | 26.2% | estimated ± 8.7 pp, medium confidence |
| ParseBench-2 | 69.3% | estimated ± 5.0 pp, medium confidence |
| CharXiv | 80.7% | estimated ± 5.6 pp, medium confidence |
| CharXiv w/o tools | 82.2% | estimated ± 1.2 pp, high confidence |
| ExploitBench | 18.0% | estimated ± 13.4 pp, low confidence |
| ExploitGym | 2.2% | estimated ± 10.6 pp, low confidence |
| AI2D_TEST | 93.9% | estimated ± 2.0 pp, medium confidence |
| C-Eval | 94.3% | estimated ± 0.9 pp, low confidence |
| Claw-Eval | 53.1% | estimated ± 2.9 pp, medium confidence |
| CountBench | 97.0% | estimated ± 2.9 pp, medium confidence |
| DynaMath | 85.5% | estimated ± 3.0 pp, medium confidence |
| ERQA | 68.1% | estimated ± 3.1 pp, high confidence |
| HMMT Feb 2026 | 89.0% | estimated ± 1.5 pp, medium confidence |
| HMMT Nov 2025 | 93.8% | estimated ± 1.5 pp, medium confidence |
| MathVision | 87.4% | estimated ± 3.0 pp, high confidence |
| MCP Atlas | 64.6% | estimated ± 7.8 pp, low confidence |
| MLVU (M-Avg) | 86.0% | estimated ± 2.2 pp, high confidence |
| MMAnswerBench | 84.7% | estimated ± 2.4 pp, medium confidence |
| MMLU-Redux | 95.3% | estimated ± 0.8 pp, medium confidence |
| MMMU-Pro | 78.1% | measured |
| NL2Repo | 40.7% | estimated ± 4.7 pp, high confidence |
| PostTrainBench v1 | 13.7% | estimated ± 3.6 pp, medium confidence |
| QwenClawBench | 56.2% | estimated ± 6.0 pp, low confidence |
| RealWorldQA | 83.7% | estimated ± 2.8 pp, high confidence |
| ScreenSpot Pro | 75.4% | estimated ± 9.6 pp, low confidence |
| SimpleVQA | 66.4% | estimated ± 7.5 pp, medium confidence |
| SuperGPQA | 75.2% | estimated ± 7.2 pp, medium confidence |
| Terminal-Bench 2.0 | 51.8% | estimated ± 5.8 pp, medium confidence |
| Toolathlon | 54.2% | estimated ± 6.3 pp, medium confidence |
| V* | 100.0% | estimated ± 7.5 pp, medium confidence |
| Video-MME (w/o subtitle) | 84.5% | estimated ± 1.3 pp, high confidence |
| Video-MME (with subtitle) | 86.8% | estimated ± 3.0 pp, high confidence |
| VideoMMMU | 85.4% | estimated ± 2.4 pp, high confidence |
| We-Math | 82.3% | estimated ± 7.4 pp, low confidence |
| AA LiveCodeBench | 87.2% | estimated ± 2.6 pp, high confidence |
| HealthBench Hard | 2.4% | estimated ± 5.1 pp, low confidence |
| MedXpertQA (MM) | 70.4% | estimated ± 2.4 pp, high confidence |
| React Native Evals | 74.7% | estimated ± 3.2 pp, high confidence |
| ResearchClawBench | 12.4% | measured |
| SWE-Rebench | 11.2% | estimated ± 8.9 pp, low confidence |
| PostTrainBench | 14.1% | estimated ± 3.7 pp, medium confidence |
| DAYJOB | 0.4% | estimated ± 6.3 pp, medium confidence |
| DAYJOB Healthcare | 0.1% | estimated ± 6.1 pp, low confidence |
| BioMysteryBench (human-solvable) | 82.2% | estimated ± 0.6 pp, low confidence |
| HealthBench (raw) | 46.1% | estimated ± 6.0 pp, low confidence |
| HealthBench (length-adjusted) | 52.5% | estimated ± 3.4 pp, low confidence |
| HealthBench Professional | 51.3% | estimated ± 4.2 pp, low confidence |
| HealthBench Professional (raw) | 56.8% | estimated ± 4.5 pp, medium confidence |
| HLE-Verified | 18.6% | estimated ± 2.0 pp, low confidence |
| LABBench2 | 80.8% | estimated ± 1.8 pp, low confidence |
| ScarfBench | 24.4% | estimated ± 9.1 pp, low confidence |
| BlenderWorld (text-only) | 70.6% | estimated ± 4.0 pp, low confidence |
| Cookie-Bench | 76.4% | estimated ± 2.5 pp, low confidence |
| STEMBench | 2.5% | estimated ± 0.2 pp, medium confidence |
| SimpleQA | 38.2% | estimated ± 4.3 pp, low confidence |
| BabyVision w/ Python | 88.8% | estimated ± 3.8 pp, low confidence |
| OpenHarmony Bench v1.0 | 51.2% | estimated ± 2.4 pp, high confidence |
| SEC-Bench Pro | 27.8% | estimated ± 12.3 pp, low confidence |
| BabyVision | 61.0% | estimated ± 9.1 pp, low confidence |
| MMVU | 77.2% | estimated ± 2.7 pp, medium confidence |
| PerceptionBench | 59.7% | estimated ± 4.9 pp, low confidence |
| MRCRv2 | 41.4% | estimated ± 11.3 pp, low confidence |
| MMLongBench-Doc | 60.6% | estimated ± 2.6 pp, low confidence |
| OCRBench V2 | 71.2% | estimated ± 3.3 pp, high confidence |
| BioBasicsBench | 22.0% | estimated ± 2.2 pp, low confidence |
| MMMU-Pro w/ Python | 81.0% | estimated ± 2.0 pp, high confidence |
| ExploitBench Internal Port | 0.0% | estimated ± 1.4 pp, low confidence |