Anthropic · model
Claude Opus 4.6 (Adaptive) benchmark scores
As of 2026-11-22, Claude Opus 4.6 (Adaptive) (Anthropic) has measured scores on 18 benchmarks and estimated scores on 101 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 89.6% | measured |
| AA-HLE | 39.9% | measured |
| AA-SciCode | 54.0% | estimated ± 3.8 pp, high confidence |
| Artificial Analysis Intelligence Index | 32.0% | measured |
| CritPt | 12.6% | measured |
| GDPval-AA | 25.6% | estimated ± 11.0 pp, low confidence |
| GPQA Diamond | 92.2% | estimated ± 1.4 pp, high confidence |
| HLE | 48.4% | estimated ± 6.1 pp, medium confidence |
| IFBench | 56.9% | estimated ± 6.6 pp, medium confidence |
| AA-LCR | 78.0% | measured |
| LiveCodeBench v6 | 88.6% | estimated ± 2.2 pp, medium confidence |
| AA-Omniscience Accuracy | 47.0% | measured |
| SciCode | 46.1% | estimated ± 2.9 pp, medium confidence |
| BrowseComp | 85.0% | estimated ± 4.2 pp, high confidence |
| HLE w/ tools | 57.0% | estimated ± 4.1 pp, medium confidence |
| IFEval | 91.5% | estimated ± 2.8 pp, high confidence |
| LongBench v2 | 64.0% | estimated ± 5.5 pp, medium confidence |
| τ²-bench results | 92.1% | measured |
| AA Coding Index | 63.2% | estimated ± 5.6 pp, medium confidence |
| AA-MMMU-Pro | 75.4% | measured |
| APEX-Agents | 33.8% | estimated ± 4.9 pp, low confidence |
| APEX-Agents-AA | 33.0% | measured |
| BioMysteryBench (human-difficult) | 45.5% | estimated ± 6.8 pp, low confidence |
| SWE-bench Verified | 80.9% | estimated ± 4.4 pp, high confidence |
| BFCL v4 | 67.5% | estimated ± 8.5 pp, low confidence |
| JobBench | 41.4% | estimated ± 8.2 pp, low confidence |
| SWE-bench Pro | 57.2% | estimated ± 2.9 pp, medium confidence |
| AA Agentic Index | 32.9% | estimated ± 12.2 pp, low confidence |
| MMLU-Pro | 87.2% | estimated ± 4.8 pp, high confidence |
| AA-IFBench | 53.1% | measured |
| FailureSensorIQ | 68.2% | estimated ± 1.7 pp, high confidence |
| GPQA | 92.3% | estimated ± 1.6 pp, high confidence |
| MMLU | 90.5% | estimated ± 0.9 pp, medium confidence |
| Vals GPQA Diamond | 89.6% | measured |
| Vals LiveCodeBench | 84.7% | measured |
| Vals MMLU-Pro | 89.1% | measured |
| MMMLU | 89.9% | estimated ± 0.7 pp, medium confidence |
| MMMU | 83.6% | estimated ± 2.7 pp, high confidence |
| Chess Puzzles | 51.8% | estimated ± 12.3 pp, low confidence |
| AA MMLU-Pro | 88.4% | estimated ± 0.7 pp, high confidence |
| Gert Labs | 54.3% | estimated ± 9.3 pp, medium confidence |
| AA Tau3 Banking | 39.2% | estimated ± 2.7 pp, medium confidence |
| ApprenticeBench | 9.2% | estimated ± 5.0 pp, low confidence |
| ARC-AGI-1 | 93.0% | measured |
| ARC-AGI-2 | 68.8% | measured |
| CursorBench 3.1 | 52.3% | estimated ± 5.8 pp, low confidence |
| CursorBench 3.2 | 57.2% | estimated ± 5.4 pp, low confidence |
| FrontierCode 1.1 Main | 27.0% | estimated ± 4.3 pp, medium confidence |
| FrontierSWE v2 | 9.4% | estimated ± 9.5 pp, medium confidence |
| Mystery Game Puzzles | 49.1% | estimated ± 14.6 pp, low confidence |
| NYT Connections | 92.1% | measured |
| OfficeQA Pro | 50.3% | estimated ± 6.8 pp, low confidence |
| ParseBench-1 | 59.6% | estimated ± 6.2 pp, medium confidence |
| Vals SWE-bench | 78.2% | measured |
| VulcanBench v3 | 83.4% | estimated ± 5.6 pp, low confidence |
| HealthBench Professional (length-adjusted, Artificial Analysis) | 55.6% | estimated ± 4.3 pp, low confidence |
| HealthBench Professional (raw, Artificial Analysis) | 59.0% | estimated ± 0.8 pp, low confidence |
| AA ITBench | 39.1% | estimated ± 6.3 pp, low confidence |
| Chartography (tools) | 80.5% | estimated ± 4.1 pp, low confidence |
| CursorBench 4.0 | 30.9% | estimated ± 3.4 pp, medium confidence |
| DeepSWE | 56.3% | estimated ± 7.2 pp, low confidence |
| HLE w/o tools | 47.4% | estimated ± 4.9 pp, high confidence |
| PostTrainBench v1.1 | 22.6% | estimated ± 3.7 pp, high confidence |
| SpireClimb | 5.8% | estimated ± 7.2 pp, medium confidence |
| StealthMark | 58.3% | estimated ± 10.7 pp, low confidence |
| ARC-AGI-3 | 5.0% | estimated ± 8.6 pp, low confidence |
| ARC-AGI-3 Provider Adapter | 2.7% | estimated ± 7.3 pp, medium confidence |
| SkillsBench official leaderboard (BenchFlow) | 55.7% | estimated ± 6.1 pp, medium confidence |
| BinaryAudit | 35.8% | estimated ± 12.1 pp, low confidence |
| OTelBench | 22.5% | estimated ± 6.6 pp, medium confidence |
| Vibe Code Bench | 53.5% | measured |
| FACTS Multimodal | 51.5% | estimated ± 4.0 pp, high confidence |
| ParseBench-2 | 65.1% | estimated ± 6.2 pp, medium confidence |
| CharXiv | 81.2% | estimated ± 6.3 pp, low confidence |
| CharXiv w/o tools | 80.4% | estimated ± 2.5 pp, high confidence |
| AI2D_TEST | 92.1% | estimated ± 2.0 pp, medium confidence |
| C-Eval | 95.1% | estimated ± 1.0 pp, low confidence |
| CountBench | 95.2% | estimated ± 2.9 pp, medium confidence |
| DynaMath | 85.3% | estimated ± 3.0 pp, medium confidence |
| ERQA | 64.4% | estimated ± 3.1 pp, high confidence |
| MathVision | 85.7% | estimated ± 3.1 pp, high confidence |
| MCP Atlas | 72.4% | estimated ± 9.1 pp, medium confidence |
| MMLU-Redux | 96.8% | estimated ± 0.8 pp, low confidence |
| MMMU-Pro | 76.6% | estimated ± 1.5 pp, high confidence |
| NL2Repo | 44.3% | estimated ± 4.7 pp, high confidence |
| PostTrainBench v1 | 20.2% | estimated ± 3.1 pp, medium confidence |
| RealWorldQA | 83.3% | estimated ± 2.9 pp, medium confidence |
| ScreenSpot Pro | 74.7% | estimated ± 10.8 pp, low confidence |
| SuperGPQA | 94.1% | estimated ± 9.3 pp, low confidence |
| Toolathlon | 50.2% | estimated ± 6.7 pp, medium confidence |
| V* | 100.0% | estimated ± 7.5 pp, medium confidence |
| Video-MME (w/o subtitle) | 83.3% | estimated ± 1.7 pp, medium confidence |
| Video-MME (with subtitle) | 85.3% | estimated ± 3.7 pp, medium confidence |
| AA LiveCodeBench | 90.6% | estimated ± 2.6 pp, medium confidence |
| HealthBench Hard | 30.3% | estimated ± 8.6 pp, low confidence |
| MedXpertQA (MM) | 72.8% | estimated ± 6.5 pp, low confidence |
| React Native Evals | 82.2% | estimated ± 3.0 pp, medium confidence |
| ResearchClawBench | 16.5% | estimated ± 2.4 pp, medium confidence |
| SWE-Rebench | 62.1% | estimated ± 6.5 pp, low confidence |
| PostTrainBench | 20.1% | estimated ± 3.7 pp, high confidence |
| BioMysteryBench (human-solvable) | 85.7% | estimated ± 0.6 pp, low confidence |
| HealthBench (raw) | 48.6% | estimated ± 6.0 pp, medium confidence |
| HealthBench (length-adjusted) | 56.3% | estimated ± 2.5 pp, low confidence |
| HealthBench Professional | 57.0% | estimated ± 4.2 pp, medium confidence |
| HealthBench Professional (raw) | 59.2% | estimated ± 4.5 pp, high confidence |
| HLE-Verified | 16.6% | estimated ± 2.0 pp, low confidence |
| LABBench2 | 77.2% | estimated ± 1.7 pp, low confidence |
| ScarfBench | 49.1% | estimated ± 12.3 pp, low confidence |
| BlenderWorld (text-only) | 82.8% | estimated ± 4.3 pp, medium confidence |
| Cookie-Bench | 89.8% | estimated ± 4.8 pp, medium confidence |
| STEMBench | 2.6% | estimated ± 0.2 pp, medium confidence |
| SimpleQA | 33.4% | estimated ± 7.8 pp, low confidence |
| BabyVision w/ Python | 87.9% | estimated ± 4.0 pp, low confidence |
| OpenHarmony Bench v1.0 | 53.7% | estimated ± 3.2 pp, high confidence |
| MRCRv2 | 85.5% | estimated ± 11.3 pp, low confidence |
| MMLongBench-Doc | 61.3% | estimated ± 2.6 pp, low confidence |
| BioBasicsBench | 33.9% | estimated ± 2.8 pp, medium confidence |
| MMMU-Pro w/ Python | 78.6% | estimated ± 2.0 pp, high confidence |
| ExploitBench Internal Port | 17.6% | estimated ± 13.8 pp, low confidence |