Anthropic · model
Claude Opus 4.7 benchmark scores
As of 2026-11-22, Claude Opus 4.7 (Anthropic) has measured scores on 55 benchmarks and estimated scores on 142 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 91.4% | measured |
| AA-HLE | 42.3% | measured |
| AA-SciCode | 55.5% | estimated ± 3.4 pp, high confidence |
| AIME26 | 99.2% | estimated ± 1.2 pp, low confidence |
| Artificial Analysis Intelligence Index | 40.7% | measured |
| CritPt | 12.0% | measured |
| GDPval-AA | 42.6% | measured |
| GPQA Diamond | 94.2% | measured |
| HLE | 54.7% | measured |
| IFBench | 59.3% | estimated ± 6.6 pp, medium confidence |
| AA-LCR | 78.7% | measured |
| LiveCodeBench v6 | 88.8% | estimated ± 2.2 pp, medium confidence |
| AA-Omniscience Accuracy | 48.9% | measured |
| SciCode | 46.9% | estimated ± 2.9 pp, medium confidence |
| Terminal-Bench 2.1 | 83.7% | estimated ± 6.7 pp, medium confidence |
| BrowseComp | 79.3% | measured |
| HLE w/ tools | 50.1% | estimated ± 4.9 pp, medium confidence |
| IFEval | 92.1% | estimated ± 2.8 pp, high confidence |
| LongBench v2 | 64.6% | estimated ± 5.5 pp, medium confidence |
| VITA-Bench | 44.6% | estimated ± 6.7 pp, low confidence |
| τ²-bench results | 88.6% | measured |
| AA Coding Index | 73.6% | measured |
| AA-MMMU-Pro | 78.8% | measured |
| APEX-Agents | 34.6% | estimated ± 9.4 pp, low confidence |
| APEX-Agents-AA | 36.1% | estimated ± 2.9 pp, medium confidence |
| BioMysteryBench (human-difficult) | 45.1% | estimated ± 6.8 pp, low confidence |
| DeepSearchQA | 91.6% | estimated ± 4.5 pp, high confidence |
| SWE-bench Verified | 87.6% | measured |
| AutomationBench | 29.2% | estimated ± 10.4 pp, low confidence |
| BFCL v4 | 76.0% | estimated ± 2.7 pp, medium confidence |
| CyberGym | 73.1% | measured |
| JobBench | 45.9% | measured |
| SkillsBench | 50.6% | estimated ± 4.1 pp, medium confidence |
| SWE-bench Pro | 64.3% | measured |
| WideResearch | 80.4% | estimated ± 1.2 pp, medium confidence |
| AA Agentic Index | 39.5% | measured |
| MMLU-Pro | 85.8% | estimated ± 2.5 pp, low confidence |
| AA-IFBench | 58.6% | measured |
| FailureSensorIQ | 66.5% | estimated ± 2.1 pp, high confidence |
| GPQA | 94.2% | measured |
| MMLU | 90.5% | estimated ± 0.9 pp, medium confidence |
| Vals GPQA Diamond | 90.2% | measured |
| Vals LiveCodeBench | 85.1% | measured |
| Vals MMLU-Pro | 89.9% | measured |
| MMMLU | 89.8% | estimated ± 0.7 pp, medium confidence |
| MMMU | 85.7% | estimated ± 2.7 pp, high confidence |
| Chess Puzzles | 41.0% | measured |
| AA MMLU-Pro | 88.7% | estimated ± 0.7 pp, high confidence |
| Gert Labs | 47.8% | measured |
| AA-AnalystAgent | 46.5% | estimated ± 3.7 pp, medium confidence |
| AA EnterpriseOps-Gym | 45.8% | estimated ± 1.2 pp, medium confidence |
| AA Harvey LAB v1.0 | 92.1% | estimated ± 1.2 pp, medium confidence |
| AA Tau3 Banking | 39.4% | estimated ± 3.3 pp, low confidence |
| AA Terminal-Bench 2.1 | 83.1% | measured |
| ApprenticeBench | 7.0% | measured |
| ARC-AGI-1 | 93.8% | estimated ± 1.7 pp, high confidence |
| ARC-AGI-2 | 75.8% | measured |
| BabaIsBench | 94.5% | estimated ± 1.2 pp, high confidence |
| CursorBench 3.1 | 55.2% | estimated ± 4.8 pp, medium confidence |
| CursorBench 3.2 | 55.6% | estimated ± 1.6 pp, low confidence |
| EBR-bench | 19.0% | measured |
| Terminal-Bench 3.0 | 11.9% | estimated ± 6.9 pp, medium confidence |
| FrontierCode 1.1 Main | 38.5% | measured |
| FrontierMath v2 (Tier 4) | 31.7% | measured |
| FrontierMath v2 (Tiers 1-3) | 70.2% | measured |
| FrontierSWE v2 | 23.0% | estimated ± 8.3 pp, medium confidence |
| Furniture Assembly | 33.3% | measured |
| MaxxBench | 48.4% | estimated ± 4.3 pp, low confidence |
| MazeBench | 0.5% | estimated ± 2.7 pp, high confidence |
| MirrorCode | 43.3% | measured |
| Mystery Game Puzzles | 28.0% | measured |
| NYT Connections | 83.0% | estimated ± 12.6 pp, low confidence |
| OfficeQA Pro | 43.6% | measured |
| OSWorld-Verified | 78.0% | measured |
| ParseBench-1 | 63.3% | measured |
| Terminal-Bench 4.1 | 17.4% | estimated ± 14.9 pp, low confidence |
| Terminal-Bench Hard | 52.4% | estimated ± 3.9 pp, high confidence |
| Vals SWE-bench | 82.0% | measured |
| Terminal-Bench 2.1 (Vals AI) | 68.5% | measured |
| VulcanBench v3 | 83.7% | estimated ± 5.6 pp, low confidence |
| CWE-Bench-AA | 46.1% | estimated ± 4.5 pp, medium confidence |
| AA Cyber Index | 24.2% | estimated ± 9.2 pp, low confidence |
| DeepsecBench-AA | 16.3% | estimated ± 8.2 pp, medium confidence |
| HealthBench Professional (length-adjusted, Artificial Analysis) | 58.2% | estimated ± 4.3 pp, medium confidence |
| HealthBench Professional (raw, Artificial Analysis) | 58.2% | estimated ± 0.8 pp, low confidence |
| AA ITBench | 46.7% | measured |
| Bug Hunt Bench | 17.3% | estimated ± 6.3 pp, low confidence |
| Chartography (tools) | 82.6% | estimated ± 4.1 pp, low confidence |
| CursorBench 4.0 | 38.1% | estimated ± 2.0 pp, high confidence |
| CWE-bench v1 | 49.3% | estimated ± 5.7 pp, medium confidence |
| DeepSWE | 62.1% | estimated ± 4.3 pp, high confidence |
| DRACO | 81.8% | estimated ± 2.0 pp, medium confidence |
| FrontierCode 2.0 | 20.0% | estimated ± 6.0 pp, medium confidence |
| FrontierPhysics | 1.2% | estimated ± 4.4 pp, high confidence |
| HLE w/o tools | 46.9% | measured |
| OSWorld 2.0 | 18.2% | measured |
| OSWorld 2.1 | 58.7% | estimated ± 3.6 pp, high confidence |
| OSWorld 2.1 Offline | 37.0% | estimated ± 1.9 pp, low confidence |
| PostTrainBench v1.1 | 28.6% | measured |
| ProgramBench | 0.5% | estimated ± 4.8 pp, low confidence |
| SpireClimb | 0.1% | estimated ± 9.3 pp, medium confidence |
| StealthMark | 45.6% | estimated ± 6.5 pp, medium confidence |
| SWE-bench Multimodal | 45.2% | estimated ± 5.6 pp, low confidence |
| Toolathlon-Verified | 72.3% | estimated ± 1.0 pp, low confidence |
| VulcanBench Frontier v4 | 80.6% | estimated ± 2.8 pp, medium confidence |
| AA AutomationBench | 53.0% | estimated ± 9.9 pp, low confidence |
| GDP.pdf | 16.5% | estimated ± 4.1 pp, medium confidence |
| AA Terminal-Bench 4.0 | 25.2% | estimated ± 13.7 pp, low confidence |
| ARC-AGI-3 | 0.2% | measured |
| ARC-AGI-3 Provider Adapter | 1.0% | estimated ± 4.4 pp, high confidence |
| CWE-bench | 45.6% | estimated ± 6.2 pp, medium confidence |
| MKA | 35.1% | estimated ± 4.6 pp, high confidence |
| SkillsBench official leaderboard (BenchFlow) | 61.2% | measured |
| BinaryAudit | 47.2% | estimated ± 12.0 pp, low confidence |
| CompileBench | 100.0% | estimated ± 10.0 pp, low confidence |
| OTelBench | 30.3% | estimated ± 4.1 pp, low confidence |
| Vibe Code Bench | 71.0% | measured |
| FACTS Multimodal | 43.9% | estimated ± 5.6 pp, medium confidence |
| Hermes Bench | 67.2% | estimated ± 2.9 pp, high confidence |
| Hermes Index | 31.2% | estimated ± 5.7 pp, low confidence |
| ParseBench-2 | 69.0% | estimated ± 3.5 pp, high confidence |
| CharXiv | 91.0% | measured |
| CharXiv w/o tools | 82.1% | measured |
| ExploitBench | 32.4% | estimated ± 6.5 pp, low confidence |
| ExploitGym | 3.6% | estimated ± 8.3 pp, low confidence |
| AI2D_TEST | 93.5% | estimated ± 0.9 pp, low confidence |
| C-Eval | 97.7% | estimated ± 1.0 pp, low confidence |
| Claw-Eval | 73.3% | estimated ± 3.2 pp, medium confidence |
| CountBench | 97.4% | estimated ± 2.9 pp, medium confidence |
| DeepPlanning | 48.8% | estimated ± 13.1 pp, low confidence |
| DynaMath | 88.2% | estimated ± 0.8 pp, low confidence |
| ERQA | 66.3% | estimated ± 2.9 pp, medium confidence |
| HMMT Feb 2026 | 95.4% | estimated ± 1.5 pp, medium confidence |
| HMMT Nov 2025 | 95.5% | estimated ± 1.5 pp, low confidence |
| MathVision | 89.5% | estimated ± 3.1 pp, high confidence |
| MCP Atlas | 77.3% | measured |
| MLVU (M-Avg) | 89.3% | estimated ± 1.5 pp, high confidence |
| MMAnswerBench | 91.7% | estimated ± 2.4 pp, low confidence |
| MMLU-Redux | 97.1% | estimated ± 0.8 pp, low confidence |
| MMMU-Pro | 79.6% | estimated ± 1.5 pp, high confidence |
| NL2Repo | 46.9% | estimated ± 4.7 pp, high confidence |
| PostTrainBench v1 | 28.6% | measured |
| QwenClawBench | 63.9% | estimated ± 2.6 pp, medium confidence |
| RealWorldQA | 87.7% | estimated ± 1.5 pp, high confidence |
| ScreenSpot Pro | 86.1% | estimated ± 6.8 pp, medium confidence |
| SuperGPQA | 93.7% | estimated ± 7.2 pp, low confidence |
| Terminal-Bench 2.0 | 69.4% | measured |
| Toolathlon | 53.0% | estimated ± 1.7 pp, medium confidence |
| V* | 100.0% | estimated ± 7.5 pp, medium confidence |
| Video-MME (w/o subtitle) | 85.8% | estimated ± 1.7 pp, medium confidence |
| Video-MME (with subtitle) | 91.1% | estimated ± 2.9 pp, high confidence |
| VideoMMMU | 87.5% | estimated ± 1.5 pp, medium confidence |
| We-Math | 84.6% | estimated ± 7.9 pp, low confidence |
| AA LiveCodeBench | 92.8% | estimated ± 2.6 pp, medium confidence |
| HealthBench Hard | 36.5% | estimated ± 5.1 pp, low confidence |
| LiveCodeBench Pro | 87.0% | estimated ± 6.0 pp, low confidence |
| MedXpertQA (MM) | 79.4% | estimated ± 6.4 pp, low confidence |
| MedXpertQA (Text) | 68.9% | estimated ± 4.5 pp, medium confidence |
| MMLU-Pro (Arcee) | 88.2% | estimated ± 3.4 pp, low confidence |
| React Native Evals | 82.8% | measured |
| ResearchClawBench | 20.7% | measured |
| SWE-Rebench | 61.8% | estimated ± 3.6 pp, low confidence |
| PostTrainBench | 24.3% | measured |
| DAYJOB | 1.4% | estimated ± 2.7 pp, high confidence |
| DAYJOB Healthcare | 0.0% | estimated ± 4.0 pp, high confidence |
| Finance Agent v2 | 55.6% | estimated ± 2.4 pp, medium confidence |
| BioMysteryBench (human-solvable) | 85.1% | estimated ± 0.6 pp, low confidence |
| FrontierCode 1.1 Extended | 57.4% | estimated ± 3.0 pp, medium confidence |
| HealthBench (raw) | 50.8% | estimated ± 6.0 pp, medium confidence |
| HealthBench (length-adjusted) | 53.8% | estimated ± 0.9 pp, medium confidence |
| HealthBench Professional | 56.8% | estimated ± 3.6 pp, high confidence |
| HealthBench Professional (raw) | 61.3% | estimated ± 4.5 pp, high confidence |
| HLE-Verified | 48.2% | estimated ± 1.3 pp, low confidence |
| LABBench2 | 83.0% | estimated ± 1.8 pp, low confidence |
| ScarfBench | 57.2% | estimated ± 5.5 pp, low confidence |
| BlenderWorld (text-only) | 70.6% | estimated ± 2.8 pp, medium confidence |
| Cookie-Bench | 78.5% | estimated ± 1.3 pp, medium confidence |
| STEMBench | 2.6% | estimated ± 0.2 pp, medium confidence |
| SimpleQA | 34.3% | estimated ± 7.8 pp, low confidence |
| LiveCodeBench | 83.7% | estimated ± 14.6 pp, low confidence |
| Agents' Last Exam | 24.9% | estimated ± 3.3 pp, low confidence |
| BabyVision w/ Python | 88.2% | estimated ± 3.7 pp, low confidence |
| OpenHarmony Bench v1.0 | 58.1% | estimated ± 2.8 pp, high confidence |
| SEC-Bench Pro | 36.9% | estimated ± 7.5 pp, low confidence |
| BabyVision | 55.2% | estimated ± 7.4 pp, low confidence |
| MMVU | 81.1% | estimated ± 3.6 pp, medium confidence |
| PerceptionBench | 55.3% | estimated ± 3.8 pp, medium confidence |
| ZeroBench | 20.4% | estimated ± 7.1 pp, low confidence |
| MRCRv2 | 85.6% | estimated ± 11.3 pp, low confidence |
| LVBench | 74.0% | estimated ± 8.2 pp, low confidence |
| MMLongBench-Doc | 61.4% | estimated ± 2.6 pp, low confidence |
| BioBasicsBench | 41.9% | estimated ± 2.2 pp, medium confidence |
| LifeVal | 70.7% | estimated ± 1.6 pp, medium confidence |
| MMMU-Pro w/ Python | 81.6% | estimated ± 2.0 pp, high confidence |
| ExploitBench Internal Port | 0.5% | estimated ± 1.4 pp, medium confidence |
| MathVision w/ Python | 96.3% | estimated ± 1.0 pp, low confidence |
| PinchBench | 90.2% | estimated ± 3.6 pp, medium confidence |