Anthropic · model
Claude Opus 5.5 benchmark scores
As of 2026-10-07, Claude Opus 5.5 (Anthropic) has measured scores on 40 benchmarks and estimated scores on 69 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 96.2% | estimated ± 2.1 pp, low confidence |
| AA-HLE | 61.4% | measured |
| AA-SciCode | 66.9% | measured |
| Artificial Analysis Intelligence Index | 57.6% | measured |
| CritPt | 31.7% | measured |
| GDPval-AA | 68.3% | measured |
| GPQA Diamond | 96.0% | estimated ± 1.0 pp, low confidence |
| HLE | 67.6% | estimated ± 1.9 pp, high confidence |
| AA-LCR | 84.7% | measured |
| LiveCodeBench v6 | 100.0% | estimated ± 2.1 pp, medium confidence |
| AA-Omniscience Accuracy | 66.2% | measured |
| SciCode | 66.1% | estimated ± 3.2 pp, medium confidence |
| Terminal-Bench 2.1 | 89.8% | estimated ± 3.9 pp, medium confidence |
| BrowseComp | 86.3% | estimated ± 2.2 pp, high confidence |
| HLE w/ tools | 67.7% | measured |
| LongBench v2 | 61.5% | estimated ± 5.9 pp, low confidence |
| τ²-bench results | 99.9% | estimated ± 4.8 pp, low confidence |
| AA Coding Index | 83.9% | estimated ± 1.0 pp, medium confidence |
| AA-MMMU-Pro | 87.7% | measured |
| APEX-Agents | 52.3% | estimated ± 10.0 pp, low confidence |
| APEX-Agents-AA | 38.4% | estimated ± 2.9 pp, medium confidence |
| BioMysteryBench (human-difficult) | 50.0% | measured |
| DeepSearchQA | 90.5% | estimated ± 8.4 pp, low confidence |
| SWE-bench Verified | 96.5% | estimated ± 0.8 pp, low confidence |
| AutomationBench | 40.0% | measured |
| BFCL v4 | 78.6% | estimated ± 2.6 pp, medium confidence |
| CyberGym | 94.9% | estimated ± 2.4 pp, low confidence |
| JobBench | 65.2% | estimated ± 3.4 pp, low confidence |
| SWE-bench Pro | 89.9% | measured |
| WideResearch | 87.9% | estimated ± 3.4 pp, low confidence |
| AA Agentic Index | 61.3% | estimated ± 2.8 pp, medium confidence |
| MMLU-Pro | 86.1% | estimated ± 2.5 pp, low confidence |
| GPQA | 96.0% | estimated ± 1.0 pp, low confidence |
| Gert Labs | 75.8% | estimated ± 3.7 pp, low confidence |
| AA-AnalystAgent | 56.3% | measured |
| AA EnterpriseOps-Gym | 55.6% | estimated ± 1.8 pp, low confidence |
| AA Harvey LAB | 91.2% | measured |
| ARC-AGI-1 | 97.5% | measured |
| ARC-AGI-2 | 91.7% | measured |
| CursorBench 3.1 | 85.7% | estimated ± 3.4 pp, low confidence |
| CursorBench 3.2 | 70.5% | estimated ± 1.6 pp, low confidence |
| FrontierCode 1.1 Main | 54.4% | measured |
| FrontierSWE v2 | 62.3% | measured |
| OfficeQA Pro | 67.7% | measured |
| OSWorld-Verified | 100.0% | estimated ± 3.0 pp, medium confidence |
| Vals GPQA Diamond | 98.2% | estimated ± 1.5 pp, low confidence |
| Vals LiveCodeBench | 89.3% | estimated ± 1.9 pp, medium confidence |
| Vals MMLU-Pro | 94.0% | estimated ± 1.5 pp, low confidence |
| Vals SWE-bench | 96.9% | estimated ± 1.4 pp, medium confidence |
| Terminal-Bench 2.1 (Vals AI) | 94.5% | estimated ± 4.6 pp, low confidence |
| VulcanBench v3 | 87.0% | estimated ± 0.7 pp, low confidence |
| AA AutomationBench | 69.5% | measured |
| GDP.pdf | 26.2% | measured |
| AA ITBench | 38.2% | measured |
| AA Tau3 Banking | 49.3% | estimated ± 3.0 pp, low confidence |
| AA Terminal-Bench 2.1 | 86.8% | estimated ± 2.6 pp, medium confidence |
| AA Terminal-Bench 4.0 | 59.6% | measured |
| CursorBench 4.0 | 57.8% | measured |
| CWE-bench v1 | 67.0% | measured |
| DeepSWE | 74.2% | measured |
| HLE w/o tools | 64.4% | measured |
| OSWorld 2.0 | 48.7% | measured |
| PostTrainBench v1.1 | 49.3% | measured |
| ProgramBench | 91.2% | measured |
| SWE Multilingual | 93.9% | measured |
| Terminal-Bench 4.0 | 66.4% | measured |
| Terminal-Bench-Science 0.1 | 58.7% | measured |
| Toolathlon-Verified | 77.8% | measured |
| Vibe Code Bench | 81.7% | estimated ± 5.6 pp, low confidence |
| CharXiv | 86.0% | estimated ± 7.5 pp, low confidence |
| CharXiv w/o tools | 87.2% | estimated ± 2.4 pp, medium confidence |
| ExploitGym | 24.0% | estimated ± 5.0 pp, low confidence |
| C-Eval | 97.1% | estimated ± 1.0 pp, low confidence |
| MathVision | 89.7% | estimated ± 4.9 pp, medium confidence |
| MCP Atlas | 87.5% | estimated ± 2.7 pp, low confidence |
| MMLU-ProX | 91.7% | estimated ± 1.7 pp, low confidence |
| MMLU-Redux | 98.1% | estimated ± 1.1 pp, low confidence |
| MMMU-Pro | 82.5% | estimated ± 4.4 pp, low confidence |
| NL2Repo | 81.4% | estimated ± 6.4 pp, low confidence |
| NOVA-63 | 64.5% | estimated ± 1.6 pp, low confidence |
| QwenClawBench | 100.0% | estimated ± 6.0 pp, low confidence |
| ScreenSpot Pro | 87.8% | estimated ± 10.7 pp, low confidence |
| SuperGPQA | 90.7% | estimated ± 7.5 pp, low confidence |
| Toolathlon | 55.9% | estimated ± 6.3 pp, low confidence |
| ERQA | 68.9% | estimated ± 3.9 pp, medium confidence |
| HealthBench Hard | 35.2% | estimated ± 7.7 pp, low confidence |
| LiveCodeBench Pro | 100.0% | estimated ± 6.0 pp, low confidence |
| MedXpertQA (MM) | 77.9% | estimated ± 6.5 pp, low confidence |
| MedXpertQA (Text) | 78.7% | estimated ± 8.9 pp, low confidence |
| MMLU-Pro (Arcee) | 100.0% | estimated ± 4.0 pp, low confidence |
| React Native Evals | 83.8% | estimated ± 4.7 pp, low confidence |
| ResearchClawBench | 20.6% | estimated ± 2.2 pp, low confidence |
| ARC-AGI-3 | 73.3% | estimated ± 3.4 pp, medium confidence |
| BioMysteryBench (human-solvable) | 89.3% | measured |
| DRACO | 89.6% | estimated ± 3.0 pp, low confidence |
| FrontierCode 1.1 Extended | 63.6% | measured |
| HealthBench (raw) | 68.1% | measured |
| HealthBench (length-adjusted) | 60.6% | measured |
| HealthBench Professional | 65.6% | measured |
| HealthBench Professional (raw) | 77.1% | measured |
| HLE-Verified | 54.6% | estimated ± 0.7 pp, medium confidence |
| LABBench2 | 84.1% | estimated ± 1.3 pp, medium confidence |
| MMMU | 94.0% | estimated ± 0.9 pp, medium confidence |
| Agents' Last Exam | 50.8% | estimated ± 6.3 pp, medium confidence |
| OpenHarmony Bench | 58.8% | estimated ± 2.4 pp, medium confidence |
| MMLU | 96.5% | estimated ± 1.1 pp, low confidence |
| MMMU-Pro w/ Python | 86.2% | estimated ± 1.6 pp, medium confidence |
| RealWorldQA | 91.4% | estimated ± 1.2 pp, low confidence |
| OmniDocBench 1.5 | 92.0% | estimated ± 6.4 pp, low confidence |