Anthropic · model
Claude Opus 4.6 benchmark scores
As of 2026-10-07, Claude Opus 4.6 (Anthropic) has measured scores on 41 benchmarks and estimated scores on 76 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 84.0% | measured |
| AA-HLE | 19.1% | measured |
| AA-SciCode | 46.9% | estimated ± 3.8 pp, high confidence |
| AIME26 | 100.0% | estimated ± 0.6 pp, low confidence |
| Artificial Analysis Intelligence Index | 26.4% | measured |
| CritPt | 2.8% | measured |
| GDPval-AA | 35.8% | estimated ± 5.1 pp, medium confidence |
| GPQA Diamond | 89.2% | measured |
| HLE | 53.0% | measured |
| IFBench | 56.4% | estimated ± 6.9 pp, medium confidence |
| AA-LCR | 67.0% | measured |
| LiveCodeBench v6 | 86.2% | estimated ± 2.1 pp, high confidence |
| AA-Omniscience Accuracy | 45.8% | measured |
| SciCode | 48.9% | estimated ± 3.8 pp, high confidence |
| BrowseComp | 83.7% | measured |
| HLE w/ tools | 57.2% | estimated ± 4.6 pp, high confidence |
| IFEval | 90.1% | estimated ± 1.3 pp, high confidence |
| LongBench v2 | 61.5% | estimated ± 5.9 pp, medium confidence |
| VITA-Bench | 45.4% | estimated ± 9.9 pp, low confidence |
| τ²-bench results | 84.8% | measured |
| AA Coding Index | 55.9% | estimated ± 3.3 pp, medium confidence |
| AA-MMMU-Pro | 72.5% | measured |
| APEX-Agents-AA | 33.8% | estimated ± 3.5 pp, low confidence |
| BioMysteryBench (human-difficult) | 43.8% | estimated ± 5.9 pp, low confidence |
| DeepSearchQA | 73.7% | measured |
| SWE-bench Verified | 80.8% | measured |
| AutomationBench | 28.4% | estimated ± 5.8 pp, low confidence |
| BFCL v4 | 85.2% | estimated ± 8.6 pp, low confidence |
| CyberGym | 66.6% | measured |
| JobBench | 36.7% | measured |
| SWE-bench Pro | 53.4% | measured |
| τ³-bench results | 68.9% | estimated ± 8.9 pp, medium confidence |
| WideResearch | 78.4% | estimated ± 1.2 pp, medium confidence |
| AA Agentic Index | 23.7% | estimated ± 5.1 pp, medium confidence |
| MMLU-Pro | 82.0% | measured |
| FrontierMath v2 (Tier 4) | 22.9% | measured |
| FrontierMath v2 (Tiers 1-3) | 40.7% | measured |
| GPQA | 91.3% | measured |
| AA-IFBench | 44.6% | measured |
| Gert Labs | 61.9% | measured |
| AA-AnalystAgent | 32.5% | estimated ± 12.5 pp, low confidence |
| AA EnterpriseOps-Gym | 29.5% | estimated ± 3.6 pp, medium confidence |
| AA Harvey LAB | 0.0% | estimated ± 2.0 pp, low confidence |
| ApprenticeBench | 5.0% | measured |
| ARC-AGI-1 | 87.9% | estimated ± 2.7 pp, high confidence |
| ARC-AGI-2 | 52.6% | estimated ± 10.1 pp, low confidence |
| CursorBench 3.1 | 50.1% | estimated ± 5.6 pp, low confidence |
| CursorBench 3.2 | 42.4% | estimated ± 1.6 pp, low confidence |
| FrontierCode 1.1 Main | 26.9% | measured |
| FrontierSWE v2 | 3.4% | estimated ± 9.0 pp, low confidence |
| OfficeQA Pro | 50.3% | estimated ± 7.2 pp, low confidence |
| OSWorld-Verified | 72.7% | measured |
| Vals GPQA Diamond | 88.4% | estimated ± 2.0 pp, high confidence |
| Vals LiveCodeBench | 81.8% | estimated ± 1.9 pp, high confidence |
| Vals MMLU-Pro | 87.5% | estimated ± 1.2 pp, high confidence |
| Vals SWE-bench | 81.6% | estimated ± 1.4 pp, medium confidence |
| Terminal-Bench 2.1 (Vals AI) | 57.4% | estimated ± 6.3 pp, medium confidence |
| AA AutomationBench | 49.8% | estimated ± 5.0 pp, medium confidence |
| GDP.pdf | 21.8% | estimated ± 4.4 pp, low confidence |
| AA ITBench | 53.3% | estimated ± 3.6 pp, low confidence |
| AA Tau3 Banking | 31.3% | estimated ± 10.2 pp, low confidence |
| CursorBench 4.0 | 30.4% | estimated ± 3.6 pp, medium confidence |
| DeepSWE | 21.8% | estimated ± 3.4 pp, low confidence |
| HLE w/o tools | 40.0% | measured |
| OSWorld 2.0 | 6.6% | estimated ± 3.6 pp, high confidence |
| PostTrainBench v1.1 | 21.8% | estimated ± 4.7 pp, medium confidence |
| Toolathlon-Verified | 73.2% | estimated ± 1.3 pp, low confidence |
| Vibe Code Bench | 57.6% | measured |
| CharXiv | 85.7% | estimated ± 3.9 pp, high confidence |
| CharXiv w/o tools | 81.5% | estimated ± 1.8 pp, medium confidence |
| ExploitGym | 11.4% | estimated ± 6.1 pp, low confidence |
| C-Eval | 100.0% | estimated ± 0.7 pp, low confidence |
| Claw-Eval | 70.4% | measured |
| HMMT Nov 2025 | 93.2% | estimated ± 2.6 pp, low confidence |
| MathVision | 86.2% | estimated ± 1.4 pp, low confidence |
| MCP Atlas | 61.9% | estimated ± 8.2 pp, medium confidence |
| MCP-Tasks | 93.7% | estimated ± 6.4 pp, low confidence |
| MMAnswerBench | 87.1% | estimated ± 1.3 pp, low confidence |
| MMLU-Redux | 91.4% | estimated ± 0.7 pp, high confidence |
| MMMU-Pro | 77.3% | measured |
| NL2Repo | 49.0% | estimated ± 6.1 pp, medium confidence |
| ScreenSpot Pro | 83.1% | measured |
| SuperGPQA | 95.0% | measured |
| Terminal-Bench 2.0 | 65.4% | measured |
| Toolathlon | 52.5% | estimated ± 1.7 pp, low confidence |
| VideoMMMU | 84.5% | estimated ± 1.0 pp, high confidence |
| ERQA | 51.6% | measured |
| HealthBench Hard | 14.8% | measured |
| LiveCodeBench Pro | 70.7% | measured |
| MedXpertQA (MM) | 64.8% | measured |
| MedXpertQA (Text) | 52.1% | measured |
| MMLU-Pro (Arcee) | 89.1% | measured |
| React Native Evals | 84.1% | measured |
| ResearchClawBench | 19.9% | measured |
| SWE-Rebench | 65.3% | measured |
| ARC-AGI-3 | 0.0% | estimated ± 2.1 pp, medium confidence |
| FrontierMath (legacy) | 41.3% | estimated ± 0.7 pp, low confidence |
| BioMysteryBench (human-solvable) | 81.7% | estimated ± 0.6 pp, low confidence |
| FrontierCode 1.1 Extended | 53.7% | estimated ± 3.0 pp, low confidence |
| HealthBench (raw) | 26.9% | estimated ± 6.4 pp, low confidence |
| HealthBench (length-adjusted) | 53.9% | estimated ± 3.7 pp, low confidence |
| HealthBench Professional | 52.7% | estimated ± 4.8 pp, low confidence |
| HealthBench Professional (raw) | 39.0% | estimated ± 5.0 pp, low confidence |
| HLE-Verified | 15.3% | estimated ± 2.0 pp, low confidence |
| LABBench2 | 68.2% | estimated ± 1.8 pp, low confidence |
| MMMU | 81.0% | estimated ± 0.9 pp, high confidence |
| LiveCodeBench | 87.0% | estimated ± 10.2 pp, low confidence |
| Agents' Last Exam | 24.0% | estimated ± 2.1 pp, medium confidence |
| OpenHarmony Bench | 53.3% | estimated ± 4.9 pp, low confidence |
| SimpleVQA | 72.8% | estimated ± 5.7 pp, low confidence |
| ZeroBench | 66.4% | estimated ± 9.3 pp, low confidence |
| MMMLU | 89.5% | estimated ± 1.1 pp, medium confidence |
| MMLU | 94.4% | estimated ± 1.1 pp, low confidence |
| MMMU-Pro w/ Python | 78.7% | estimated ± 0.6 pp, high confidence |
| RealWorldQA | 73.8% | estimated ± 1.1 pp, low confidence |
| Video-MME (with subtitle) | 87.2% | estimated ± 1.8 pp, medium confidence |
| OmniDocBench 1.5 | 91.7% | estimated ± 5.9 pp, low confidence |