Anthropic · model
Claude 4 Opus Thinking benchmark scores
As of 2026-10-29, Claude 4 Opus Thinking (Anthropic) has measured scores on 11 benchmarks and estimated scores on 48 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 79.6% | measured |
| AA-HLE | 12.3% | measured |
| AIME26 | 76.5% | estimated ± 4.6 pp, medium confidence |
| CritPt | 0.5% | estimated ± 2.4 pp, high confidence |
| GDPval-AA | 8.7% | estimated ± 10.9 pp, low confidence |
| GPQA Diamond | 79.7% | estimated ± 1.2 pp, high confidence |
| HLE | 16.4% | estimated ± 6.9 pp, medium confidence |
| IFBench | 57.1% | estimated ± 6.6 pp, medium confidence |
| AA-LCR | 69.3% | measured |
| HLE w/ tools | 23.1% | estimated ± 5.5 pp, low confidence |
| IFEval | 91.6% | estimated ± 2.8 pp, high confidence |
| LongBench v2 | 60.1% | estimated ± 5.8 pp, low confidence |
| τ²-bench results | 73.4% | measured |
| AA-MMMU-Pro | 64.3% | estimated ± 3.0 pp, high confidence |
| BioMysteryBench (human-difficult) | 19.4% | estimated ± 8.1 pp, low confidence |
| BFCL v4 | 57.8% | estimated ± 8.5 pp, low confidence |
| AA Agentic Index | 17.2% | estimated ± 12.3 pp, low confidence |
| MMLU-Pro | 86.0% | estimated ± 4.1 pp, high confidence |
| AA-IFBench | 53.7% | measured |
| AIME 2024 | 83.6% | estimated ± 2.6 pp, high confidence |
| Artificial Analysis Intelligence Index | 20.6% | measured |
| GPQA | 79.6% | measured |
| MMLU | 93.6% | estimated ± 2.7 pp, medium confidence |
| AA-Omniscience Accuracy | 40.4% | estimated ± 4.5 pp, high confidence |
| MATH-500 | 96.6% | estimated ± 2.1 pp, high confidence |
| MMMLU | 88.8% | measured |
| MMMU | 76.5% | measured |
| AA MMLU-Pro | 87.3% | measured |
| AIME 2025 | 75.5% | measured |
| ARC-AGI-1 | 47.0% | estimated ± 7.0 pp, low confidence |
| ARC-AGI-2 | 0.0% | estimated ± 11.6 pp, low confidence |
| Vals GPQA Diamond | 69.8% | estimated ± 4.8 pp, medium confidence |
| Vals MMLU-Pro | 67.9% | estimated ± 1.2 pp, medium confidence |
| HLE w/o tools | 32.9% | estimated ± 4.6 pp, high confidence |
| ARC-AGI-3 | 1.4% | estimated ± 7.7 pp, low confidence |
| AI2D_TEST | 89.8% | estimated ± 2.0 pp, high confidence |
| C-Eval | 91.7% | estimated ± 1.0 pp, medium confidence |
| CountBench | 94.6% | estimated ± 3.2 pp, high confidence |
| DynaMath | 83.1% | estimated ± 5.8 pp, low confidence |
| ERQA | 47.1% | estimated ± 5.2 pp, low confidence |
| HMMT Feb 2025 | 59.7% | estimated ± 3.1 pp, medium confidence |
| MathVision | 75.7% | estimated ± 3.6 pp, high confidence |
| MLVU (M-Avg) | 83.7% | estimated ± 2.4 pp, high confidence |
| MMLU-Redux | 94.5% | estimated ± 0.5 pp, medium confidence |
| MMMU-Pro | 65.9% | estimated ± 2.3 pp, high confidence |
| RealWorldQA | 81.0% | estimated ± 3.6 pp, high confidence |
| ScreenSpot Pro | 53.0% | estimated ± 7.3 pp, low confidence |
| SimpleVQA | 57.4% | estimated ± 7.1 pp, medium confidence |
| SuperGPQA | 61.1% | estimated ± 8.2 pp, low confidence |
| Video-MME (w/o subtitle) | 77.2% | estimated ± 1.3 pp, high confidence |
| Video-MME (with subtitle) | 81.9% | estimated ± 3.2 pp, high confidence |
| VideoMMMU | 72.5% | estimated ± 3.8 pp, high confidence |
| We-Math | 78.1% | estimated ± 7.2 pp, low confidence |
| HealthBench (length-adjusted) | 53.9% | estimated ± 3.3 pp, medium confidence |
| HealthBench Professional (raw) | 55.7% | estimated ± 8.6 pp, low confidence |
| SimpleQA | 28.4% | estimated ± 7.8 pp, medium confidence |
| MMVU | 70.6% | estimated ± 3.8 pp, medium confidence |
| MRCRv2 | 71.1% | estimated ± 11.3 pp, low confidence |
| MMLongBench-Doc | 60.9% | estimated ± 2.6 pp, low confidence |