OpenAI · model
GPT-6.2 Sol benchmark scores
As of 2026-11-18, GPT-6.2 Sol (OpenAI) has measured scores on 25 benchmarks and estimated scores on 72 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 96.2% | estimated ± 2.1 pp, low confidence |
| AA-HLE | 56.1% | estimated ± 3.9 pp, high confidence |
| AA-SciCode | 61.4% | estimated ± 3.5 pp, medium confidence |
| Artificial Analysis Intelligence Index | 57.3% | measured |
| CritPt | 32.1% | estimated ± 4.0 pp, high confidence |
| GDPval-AA | 61.7% | estimated ± 6.9 pp, low confidence |
| GPQA Diamond | 96.0% | measured |
| HLE | 70.0% | estimated ± 2.2 pp, high confidence |
| AA-Omniscience Accuracy | 61.6% | measured |
| Terminal-Bench 2.1 | 89.1% | estimated ± 5.4 pp, low confidence |
| BrowseComp | 91.4% | estimated ± 2.9 pp, low confidence |
| HLE w/ tools | 64.6% | estimated ± 5.1 pp, low confidence |
| AA Coding Index | 79.0% | estimated ± 2.3 pp, low confidence |
| BioMysteryBench (human-difficult) | 49.9% | estimated ± 7.8 pp, low confidence |
| AutomationBench | 47.1% | estimated ± 6.1 pp, low confidence |
| JobBench | 73.7% | estimated ± 11.3 pp, low confidence |
| SWE-bench Pro | 85.5% | estimated ± 8.4 pp, medium confidence |
| AA Agentic Index | 52.5% | estimated ± 12.7 pp, low confidence |
| MMLU-Pro | 85.5% | estimated ± 5.1 pp, low confidence |
| GPQA | 96.0% | measured |
| MMLU | 88.1% | estimated ± 4.3 pp, medium confidence |
| Vals GPQA Diamond | 98.8% | estimated ± 1.2 pp, low confidence |
| Vals MMLU-Pro | 89.8% | estimated ± 1.6 pp, medium confidence |
| MMMLU | 91.6% | estimated ± 1.1 pp, low confidence |
| Chess Puzzles | 76.8% | estimated ± 5.3 pp, low confidence |
| AA MMLU-Pro | 90.5% | estimated ± 0.8 pp, medium confidence |
| AA-AnalystAgent | 53.8% | measured |
| AA EnterpriseOps-Gym | 28.9% | measured |
| AA Tau3 Banking | 100.0% | estimated ± 2.4 pp, low confidence |
| ARC-AGI-1 | 97.5% | measured |
| ARC-AGI-2 | 85.8% | measured |
| CursorBench 3.2 | 82.3% | estimated ± 4.1 pp, low confidence |
| EBR-bench | 89.4% | estimated ± 8.6 pp, low confidence |
| FrontierCode 1.1 Main | 51.7% | estimated ± 3.9 pp, medium confidence |
| FrontierSWE v2 | 60.2% | estimated ± 8.3 pp, medium confidence |
| Furniture Assembly | 85.1% | estimated ± 4.8 pp, medium confidence |
| MazeBench | 9.5% | estimated ± 6.2 pp, low confidence |
| MirrorCode | 68.2% | estimated ± 14.2 pp, low confidence |
| Mystery Game Puzzles | 87.5% | estimated ± 6.4 pp, low confidence |
| NYT Connections | 100.0% | estimated ± 8.9 pp, low confidence |
| Terminal-Bench 4.1 | 63.2% | estimated ± 5.5 pp, medium confidence |
| Terminal-Bench 2.1 (Vals AI) | 82.7% | estimated ± 6.9 pp, low confidence |
| CWE-Bench-AA | 61.7% | measured |
| AA Cyber Index | 37.7% | measured |
| DeepsecBench-AA | 51.6% | measured |
| Harvey LAB-AA v1.1 | 11.1% | measured |
| HealthBench Professional (length-adjusted, Artificial Analysis) | 63.3% | estimated ± 4.3 pp, medium confidence |
| HealthBench Professional (raw, Artificial Analysis) | 70.5% | estimated ± 1.1 pp, medium confidence |
| Bug Hunt Bench | 47.2% | estimated ± 8.9 pp, medium confidence |
| CursorBench 4.0 | 57.4% | estimated ± 3.3 pp, high confidence |
| FrontierCode 2.0 | 43.8% | measured |
| FrontierPhysics | 20.6% | estimated ± 2.3 pp, high confidence |
| HLE w/o tools | 50.0% | estimated ± 4.2 pp, medium confidence |
| OSWorld 2.0 | 67.1% | estimated ± 14.7 pp, low confidence |
| OSWorld 2.1 Offline | 76.7% | estimated ± 7.9 pp, low confidence |
| PostTrainBench v1.1 | 45.7% | estimated ± 4.8 pp, high confidence |
| SpireClimb | 39.5% | estimated ± 2.4 pp, medium confidence |
| StealthMark | 68.6% | estimated ± 6.1 pp, medium confidence |
| Terminal-Bench 4.0 | 63.3% | estimated ± 6.1 pp, medium confidence |
| Terminal-Bench-Science 0.1 | 68.0% | measured |
| Toolathlon-Verified | 77.9% | estimated ± 2.8 pp, low confidence |
| VulcanBench Frontier v4 | 91.6% | estimated ± 6.0 pp, low confidence |
| AA AutomationBench | 71.3% | measured |
| GDP.pdf | 30.0% | measured |
| AA Terminal-Bench 4.0 | 60.1% | measured |
| ARC-AGI-3 | 54.9% | measured |
| ARC-AGI-3 Provider Adapter | 98.2% | measured |
| CWE-bench | 64.2% | estimated ± 6.2 pp, low confidence |
| MKA | 48.3% | estimated ± 5.3 pp, medium confidence |
| FACTS Multimodal | 52.4% | estimated ± 2.7 pp, medium confidence |
| Hermes Bench | 100.0% | estimated ± 1.8 pp, low confidence |
| Hermes Index | 92.4% | estimated ± 2.7 pp, low confidence |
| ExploitBench | 100.0% | measured |
| ExploitGym | 30.9% | estimated ± 9.8 pp, low confidence |
| C-Eval | 93.8% | estimated ± 1.2 pp, medium confidence |
| MMLU-Redux | 98.6% | estimated ± 0.8 pp, low confidence |
| SuperGPQA | 83.4% | estimated ± 8.5 pp, low confidence |
| HealthBench Hard | 34.2% | estimated ± 1.2 pp, medium confidence |
| MedXpertQA (Text) | 77.7% | estimated ± 4.5 pp, low confidence |
| MMLU-Pro (Arcee) | 88.8% | estimated ± 3.4 pp, low confidence |
| PostTrainBench | 41.9% | estimated ± 4.6 pp, high confidence |
| DAYJOB | 16.6% | estimated ± 1.1 pp, medium confidence |
| DAYJOB Healthcare | 11.2% | estimated ± 1.1 pp, low confidence |
| HealthBench (raw) | 54.1% | estimated ± 1.6 pp, high confidence |
| HealthBench (length-adjusted) | 58.4% | estimated ± 0.9 pp, medium confidence |
| HealthBench Professional | 65.9% | measured |
| HealthBench Professional (raw) | 65.5% | measured |
| HLE-Verified | 54.5% | estimated ± 0.7 pp, medium confidence |
| LABBench2 | 84.6% | estimated ± 1.7 pp, medium confidence |
| ScarfBench | 69.9% | estimated ± 5.0 pp, low confidence |
| STEMBench | 2.7% | estimated ± 0.2 pp, medium confidence |
| SimpleQA | 54.1% | estimated ± 14.6 pp, low confidence |
| Agents' Last Exam | 61.9% | estimated ± 11.4 pp, low confidence |
| SEC-Bench Pro | 88.3% | measured |
| BioBasicsBench | 66.8% | estimated ± 2.2 pp, medium confidence |
| LifeVal | 83.8% | measured |
| ExploitBench Internal Port | 50.5% | measured |