Anthropic · model
Claude Sonnet 5 benchmark scores
As of 2026-10-07, Claude Sonnet 5 (Anthropic) has measured scores on 34 benchmarks and estimated scores on 83 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 91.1% | measured |
| AA-HLE | 41.3% | measured |
| AA-SciCode | 54.3% | measured |
| Artificial Analysis Intelligence Index | 38.2% | measured |
| CritPt | 16.9% | measured |
| GDPval-AA | 48.3% | measured |
| GPQA Diamond | 90.1% | estimated ± 1.7 pp, high confidence |
| HLE | 57.4% | measured |
| AA-LCR | 82.0% | measured |
| LiveCodeBench v6 | 90.0% | estimated ± 2.1 pp, high confidence |
| AA-Omniscience Accuracy | 40.1% | measured |
| SciCode | 41.5% | estimated ± 2.9 pp, low confidence |
| Terminal-Bench 2.1 | 80.4% | measured |
| BrowseComp | 84.7% | measured |
| HLE w/ tools | 57.4% | measured |
| LongBench v2 | 61.5% | estimated ± 5.9 pp, low confidence |
| VITA-Bench | 45.6% | estimated ± 9.9 pp, low confidence |
| τ²-bench results | 88.6% | estimated ± 8.8 pp, low confidence |
| AA Coding Index | 71.6% | measured |
| AA-MMMU-Pro | 77.3% | measured |
| APEX-Agents | 40.0% | estimated ± 10.0 pp, low confidence |
| APEX-Agents-AA | 38.0% | estimated ± 3.5 pp, low confidence |
| BioMysteryBench (human-difficult) | 41.0% | estimated ± 5.9 pp, low confidence |
| DeepSearchQA | 92.5% | estimated ± 4.5 pp, high confidence |
| SWE-bench Verified | 85.2% | measured |
| AutomationBench | 28.7% | estimated ± 9.7 pp, low confidence |
| BFCL v4 | 77.3% | estimated ± 2.6 pp, medium confidence |
| CyberGym | 80.1% | estimated ± 2.4 pp, medium confidence |
| JobBench | 49.9% | estimated ± 8.6 pp, medium confidence |
| SWE-bench Pro | 63.2% | measured |
| WideResearch | 82.8% | estimated ± 3.4 pp, low confidence |
| AA Agentic Index | 44.3% | measured |
| MMLU-Pro | 85.7% | estimated ± 2.5 pp, low confidence |
| GPQA | 91.3% | estimated ± 1.6 pp, high confidence |
| Gert Labs | 69.7% | estimated ± 3.0 pp, medium confidence |
| AA-AnalystAgent | 51.5% | estimated ± 2.5 pp, medium confidence |
| AA EnterpriseOps-Gym | 46.1% | estimated ± 5.9 pp, low confidence |
| AA Harvey LAB | 92.9% | estimated ± 1.2 pp, medium confidence |
| ApprenticeBench | 16.0% | measured |
| ARC-AGI-1 | 95.4% | estimated ± 0.7 pp, low confidence |
| ARC-AGI-2 | 62.5% | estimated ± 3.1 pp, low confidence |
| CursorBench 3.1 | 54.6% | estimated ± 3.4 pp, medium confidence |
| CursorBench 3.2 | 61.5% | measured |
| Terminal-Bench 3.0 | 14.6% | measured |
| FrontierCode 1.1 Main | 42.7% | measured |
| FrontierSWE v2 | 11.1% | estimated ± 7.6 pp, low confidence |
| OfficeQA Pro | 50.4% | estimated ± 7.2 pp, medium confidence |
| OSWorld-Verified | 81.2% | measured |
| Vals GPQA Diamond | 88.9% | measured |
| Vals LiveCodeBench | 82.4% | measured |
| Vals MMLU-Pro | 87.5% | measured |
| Vals SWE-bench | 79.6% | measured |
| Terminal-Bench 2.1 (Vals AI) | 74.5% | measured |
| VulcanBench v3 | 75.2% | estimated ± 0.7 pp, low confidence |
| AA AutomationBench | 26.9% | estimated ± 5.7 pp, low confidence |
| GDP.pdf | 22.9% | estimated ± 4.4 pp, low confidence |
| AA ITBench | 49.6% | estimated ± 3.4 pp, medium confidence |
| AA Tau3 Banking | 47.4% | estimated ± 5.8 pp, medium confidence |
| AA Terminal-Bench 2.1 | 84.6% | estimated ± 3.5 pp, high confidence |
| AA Terminal-Bench 4.0 | 19.0% | estimated ± 8.9 pp, medium confidence |
| CursorBench 4.0 | 34.1% | measured |
| CWE-bench v1 | 54.7% | estimated ± 6.9 pp, medium confidence |
| DeepSWE | 65.7% | estimated ± 3.4 pp, low confidence |
| HLE w/o tools | 43.2% | measured |
| OSWorld 2.0 | 16.1% | estimated ± 3.6 pp, high confidence |
| PostTrainBench v1.1 | 25.6% | estimated ± 2.7 pp, high confidence |
| ProgramBench | 78.3% | estimated ± 6.0 pp, low confidence |
| SWE Multilingual | 78.3% | measured |
| Toolathlon-Verified | 70.9% | estimated ± 2.3 pp, medium confidence |
| Vibe Code Bench | 71.8% | estimated ± 5.6 pp, low confidence |
| CharXiv | 88.3% | measured |
| CharXiv w/o tools | 77.0% | measured |
| ExploitGym | 14.6% | estimated ± 6.1 pp, low confidence |
| C-Eval | 93.3% | estimated ± 0.8 pp, low confidence |
| Claw-Eval | 79.5% | estimated ± 7.9 pp, medium confidence |
| MathVision | 80.1% | estimated ± 2.1 pp, low confidence |
| MCP Atlas | 79.5% | estimated ± 7.3 pp, medium confidence |
| MMLU-ProX | 86.0% | estimated ± 1.7 pp, medium confidence |
| MMLU-Redux | 96.7% | estimated ± 1.1 pp, medium confidence |
| MMMU-Pro | 73.7% | estimated ± 1.4 pp, low confidence |
| NL2Repo | 45.6% | estimated ± 4.7 pp, high confidence |
| NOVA-63 | 58.5% | estimated ± 1.6 pp, low confidence |
| QwenClawBench | 100.0% | estimated ± 6.0 pp, low confidence |
| ScreenSpot Pro | 84.2% | estimated ± 7.1 pp, medium confidence |
| SuperGPQA | 81.2% | estimated ± 7.5 pp, medium confidence |
| Terminal-Bench 2.0 | 76.4% | estimated ± 5.9 pp, medium confidence |
| Toolathlon | 54.9% | estimated ± 1.7 pp, medium confidence |
| VideoMMMU | 86.7% | estimated ± 1.5 pp, high confidence |
| ERQA | 70.4% | estimated ± 3.7 pp, high confidence |
| HealthBench Hard | 32.9% | estimated ± 7.7 pp, low confidence |
| LiveCodeBench Pro | 86.2% | estimated ± 6.0 pp, low confidence |
| MedXpertQA (MM) | 78.4% | estimated ± 6.4 pp, low confidence |
| MedXpertQA (Text) | 60.7% | estimated ± 8.9 pp, low confidence |
| MMLU-Pro (Arcee) | 100.0% | estimated ± 4.0 pp, low confidence |
| React Native Evals | 80.6% | estimated ± 2.4 pp, medium confidence |
| ResearchClawBench | 20.5% | estimated ± 2.0 pp, medium confidence |
| SWE-Rebench | 61.8% | estimated ± 3.6 pp, low confidence |
| ARC-AGI-3 | 1.7% | estimated ± 3.4 pp, medium confidence |
| BioMysteryBench (human-solvable) | 88.4% | estimated ± 0.6 pp, medium confidence |
| DRACO | 81.4% | estimated ± 3.0 pp, medium confidence |
| FrontierCode 1.1 Extended | 55.1% | estimated ± 1.6 pp, medium confidence |
| HealthBench (raw) | 50.3% | estimated ± 6.4 pp, low confidence |
| HealthBench (length-adjusted) | 53.9% | estimated ± 3.7 pp, medium confidence |
| HealthBench Professional | 59.0% | estimated ± 1.3 pp, medium confidence |
| HealthBench Professional (raw) | 60.5% | estimated ± 5.0 pp, medium confidence |
| HLE-Verified | 31.0% | measured |
| LABBench2 | 80.1% | measured |
| MMMU | 84.9% | estimated ± 0.9 pp, high confidence |
| LiveCodeBench | 100.0% | estimated ± 8.9 pp, low confidence |
| OpenHarmony Bench | 58.8% | estimated ± 2.4 pp, high confidence |
| SimpleVQA | 71.0% | estimated ± 7.8 pp, low confidence |
| MMLU | 93.4% | estimated ± 1.1 pp, low confidence |
| MMMU-Pro w/ Python | 80.1% | estimated ± 1.6 pp, high confidence |
| MathVision w/ Python | 82.9% | estimated ± 0.9 pp, low confidence |
| RealWorldQA | 86.9% | estimated ± 1.1 pp, medium confidence |
| Video-MME (with subtitle) | 89.1% | estimated ± 0.9 pp, medium confidence |
| OmniDocBench 1.5 | 92.0% | estimated ± 6.4 pp, low confidence |