Anthropic · model
Claude 4.1 Opus benchmark scores
As of 2026-10-07, Claude 4.1 Opus (Anthropic) has measured scores on 3 benchmarks and estimated scores on 52 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 82.4% | estimated ± 6.8 pp, medium confidence |
| AA-HLE | 18.7% | estimated ± 4.4 pp, high confidence |
| AA-SciCode | 45.7% | estimated ± 3.9 pp, high confidence |
| Artificial Analysis Intelligence Index | 18.6% | measured |
| CritPt | 1.7% | estimated ± 3.7 pp, high confidence |
| GDPval-AA | 29.0% | estimated ± 6.2 pp, medium confidence |
| GPQA Diamond | 83.0% | estimated ± 2.4 pp, high confidence |
| HLE | 20.3% | estimated ± 9.9 pp, medium confidence |
| LiveCodeBench v6 | 85.4% | estimated ± 8.1 pp, low confidence |
| AA-Omniscience Accuracy | 24.8% | estimated ± 7.5 pp, medium confidence |
| SciCode | 45.9% | estimated ± 3.8 pp, high confidence |
| BrowseComp | 68.4% | estimated ± 6.2 pp, medium confidence |
| AA Coding Index | 49.4% | estimated ± 5.1 pp, medium confidence |
| APEX-Agents-AA | 24.6% | estimated ± 3.6 pp, medium confidence |
| BioMysteryBench (human-difficult) | 35.2% | estimated ± 8.1 pp, low confidence |
| DeepSearchQA | 76.9% | estimated ± 10.9 pp, low confidence |
| SWE-bench Verified | 74.5% | measured |
| CyberGym | 47.6% | estimated ± 11.4 pp, low confidence |
| JobBench | 21.9% | measured |
| SWE-bench Pro | 50.1% | estimated ± 3.2 pp, high confidence |
| WideResearch | 75.1% | estimated ± 1.2 pp, medium confidence |
| MMLU-Pro | 84.8% | estimated ± 4.5 pp, high confidence |
| GPQA | 83.8% | estimated ± 3.2 pp, high confidence |
| Gert Labs | 50.8% | estimated ± 7.3 pp, medium confidence |
| ApprenticeBench | 0.5% | estimated ± 5.6 pp, low confidence |
| ARC-AGI-1 | 40.2% | estimated ± 5.4 pp, low confidence |
| ARC-AGI-2 | 12.7% | estimated ± 14.2 pp, low confidence |
| FrontierCode 1.1 Main | 9.7% | estimated ± 2.7 pp, low confidence |
| OSWorld-Verified | 55.4% | estimated ± 8.7 pp, low confidence |
| Vals GPQA Diamond | 76.5% | estimated ± 7.5 pp, medium confidence |
| Vals LiveCodeBench | 75.3% | estimated ± 9.9 pp, low confidence |
| Vals MMLU-Pro | 82.7% | estimated ± 2.6 pp, medium confidence |
| Vals SWE-bench | 68.8% | estimated ± 6.8 pp, medium confidence |
| AA AutomationBench | 45.7% | estimated ± 5.0 pp, low confidence |
| GDP.pdf | 17.8% | estimated ± 5.2 pp, low confidence |
| HLE w/o tools | 23.8% | estimated ± 6.3 pp, medium confidence |
| Toolathlon-Verified | 73.2% | estimated ± 1.3 pp, low confidence |
| Vibe Code Bench | 2.5% | estimated ± 8.1 pp, medium confidence |
| C-Eval | 90.5% | estimated ± 1.0 pp, medium confidence |
| Claw-Eval | 50.7% | estimated ± 5.6 pp, low confidence |
| MCP Atlas | 44.3% | estimated ± 8.2 pp, medium confidence |
| MMLU-Redux | 93.7% | estimated ± 2.1 pp, medium confidence |
| NL2Repo | 36.2% | estimated ± 6.1 pp, medium confidence |
| SuperGPQA | 64.0% | estimated ± 10.4 pp, low confidence |
| Toolathlon | 33.1% | estimated ± 7.4 pp, low confidence |
| HealthBench Hard | 5.2% | estimated ± 7.7 pp, low confidence |
| MMLU-Pro (Arcee) | 82.7% | estimated ± 5.4 pp, low confidence |
| SWE-Rebench | 57.1% | estimated ± 3.6 pp, medium confidence |
| BioMysteryBench (human-solvable) | 27.9% | estimated ± 1.0 pp, low confidence |
| HealthBench (raw) | 26.9% | estimated ± 6.7 pp, low confidence |
| HealthBench (length-adjusted) | 53.9% | estimated ± 3.7 pp, low confidence |
| HealthBench Professional (raw) | 36.8% | estimated ± 5.0 pp, low confidence |
| LiveCodeBench | 83.0% | estimated ± 10.2 pp, low confidence |
| Agents' Last Exam | 48.9% | estimated ± 8.7 pp, low confidence |
| MMLU | 94.2% | estimated ± 1.9 pp, low confidence |