OpenAI · model
GPT-6 Sol benchmark scores
As of 2026-10-07, GPT-6 Sol (OpenAI) has measured scores on 27 benchmarks and estimated scores on 74 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 94.3% | estimated ± 1.2 pp, high confidence |
| AA-HLE | 47.9% | measured |
| AA-SciCode | 57.6% | measured |
| Artificial Analysis Intelligence Index | 47.6% | measured |
| CritPt | 30.9% | measured |
| GDPval-AA | 50.5% | measured |
| GPQA Diamond | 94.9% | estimated ± 1.0 pp, medium confidence |
| HLE | 60.5% | estimated ± 5.5 pp, low confidence |
| AA-LCR | 83.7% | measured |
| LiveCodeBench v6 | 98.2% | estimated ± 3.3 pp, medium confidence |
| AA-Omniscience Accuracy | 54.5% | measured |
| SciCode | 57.1% | estimated ± 3.2 pp, high confidence |
| Terminal-Bench 2.1 | 89.1% | estimated ± 3.9 pp, medium confidence |
| BrowseComp | 87.0% | estimated ± 1.8 pp, medium confidence |
| HLE w/ tools | 58.3% | estimated ± 4.6 pp, medium confidence |
| LongBench v2 | 61.5% | estimated ± 5.9 pp, low confidence |
| τ²-bench results | 85.4% | estimated ± 4.8 pp, medium confidence |
| AA Coding Index | 75.8% | estimated ± 2.5 pp, high confidence |
| AA-MMMU-Pro | 82.9% | measured |
| APEX-Agents | 41.5% | estimated ± 10.0 pp, low confidence |
| APEX-Agents-AA | 38.7% | estimated ± 2.9 pp, low confidence |
| BioMysteryBench (human-difficult) | 47.4% | estimated ± 5.9 pp, low confidence |
| DeepSearchQA | 90.3% | estimated ± 8.4 pp, medium confidence |
| SWE-bench Verified | 89.8% | estimated ± 5.2 pp, low confidence |
| AutomationBench | 33.2% | measured |
| BFCL v4 | 77.5% | estimated ± 2.6 pp, medium confidence |
| CyberGym | 80.8% | estimated ± 3.7 pp, high confidence |
| JobBench | 64.8% | estimated ± 4.9 pp, medium confidence |
| SWE-bench Pro | 71.0% | estimated ± 6.8 pp, medium confidence |
| WideResearch | 100.0% | estimated ± 9.1 pp, low confidence |
| AA Agentic Index | 45.6% | estimated ± 2.8 pp, high confidence |
| MMLU-Pro | 85.7% | estimated ± 3.5 pp, medium confidence |
| GPQA | 94.2% | estimated ± 0.8 pp, high confidence |
| Gert Labs | 76.4% | estimated ± 3.7 pp, low confidence |
| AA-AnalystAgent | 54.4% | estimated ± 5.5 pp, medium confidence |
| AA EnterpriseOps-Gym | 50.4% | estimated ± 4.4 pp, medium confidence |
| AA Harvey LAB | 92.7% | estimated ± 1.3 pp, medium confidence |
| ARC-AGI-1 | 95.5% | measured |
| ARC-AGI-2 | 89.6% | measured |
| CursorBench 3.1 | 62.7% | estimated ± 3.4 pp, medium confidence |
| CursorBench 3.2 | 65.8% | estimated ± 5.3 pp, medium confidence |
| FrontierCode 1.1 Main | 47.6% | estimated ± 3.7 pp, low confidence |
| FrontierSWE v2 | 35.1% | estimated ± 14.6 pp, low confidence |
| OfficeQA Pro | 67.6% | estimated ± 7.2 pp, medium confidence |
| OSWorld-Verified | 100.0% | estimated ± 3.0 pp, medium confidence |
| Vals GPQA Diamond | 96.2% | estimated ± 1.5 pp, low confidence |
| Vals MMLU-Pro | 90.3% | estimated ± 1.5 pp, low confidence |
| Vals SWE-bench | 88.8% | estimated ± 7.6 pp, low confidence |
| Terminal-Bench 2.1 (Vals AI) | 78.0% | estimated ± 8.5 pp, medium confidence |
| AA AutomationBench | 61.6% | measured |
| GDP.pdf | 25.2% | measured |
| AA ITBench | 49.4% | measured |
| AA Tau3 Banking | 46.1% | estimated ± 3.0 pp, low confidence |
| AA Terminal-Bench 2.1 | 86.8% | estimated ± 2.6 pp, high confidence |
| AA Terminal-Bench 4.0 | 43.9% | measured |
| CursorBench 4.0 | 44.1% | estimated ± 3.5 pp, high confidence |
| CWE-bench v1 | 52.0% | measured |
| DeepSWE | 68.8% | measured |
| HLE w/o tools | 52.1% | estimated ± 5.0 pp, medium confidence |
| OSWorld 2.0 | 60.5% | measured |
| PostTrainBench v1.1 | 37.6% | estimated ± 4.8 pp, high confidence |
| Terminal-Bench 4.0 | 47.4% | estimated ± 5.1 pp, medium confidence |
| Terminal-Bench-Science 0.1 | 39.5% | estimated ± 4.1 pp, low confidence |
| Toolathlon-Verified | 79.4% | estimated ± 1.0 pp, medium confidence |
| CharXiv | 86.0% | estimated ± 7.5 pp, low confidence |
| CharXiv w/o tools | 84.4% | estimated ± 2.4 pp, high confidence |
| ExploitBench | 61.1% | estimated ± 12.2 pp, low confidence |
| ExploitGym | 22.1% | measured |
| C-Eval | 96.4% | estimated ± 1.0 pp, low confidence |
| MathVision | 89.7% | estimated ± 4.9 pp, medium confidence |
| MCP Atlas | 85.8% | estimated ± 2.7 pp, low confidence |
| MMLU-Redux | 97.6% | estimated ± 1.1 pp, low confidence |
| MMMU-Pro | 80.5% | estimated ± 4.4 pp, medium confidence |
| NL2Repo | 61.9% | estimated ± 6.5 pp, medium confidence |
| QwenClawBench | 100.0% | estimated ± 6.0 pp, low confidence |
| ScreenSpot Pro | 82.4% | estimated ± 10.7 pp, low confidence |
| SuperGPQA | 87.3% | estimated ± 7.5 pp, low confidence |
| Toolathlon | 55.9% | estimated ± 6.3 pp, medium confidence |
| ERQA | 68.8% | estimated ± 3.9 pp, medium confidence |
| HealthBench Hard | 30.1% | measured |
| MedXpertQA (MM) | 77.6% | estimated ± 6.5 pp, low confidence |
| MMLU-Pro (Arcee) | 100.0% | estimated ± 4.0 pp, low confidence |
| React Native Evals | 79.9% | estimated ± 4.7 pp, medium confidence |
| ResearchClawBench | 19.5% | estimated ± 2.2 pp, low confidence |
| ARC-AGI-3 | 4.6% | measured |
| BioMysteryBench (human-solvable) | 89.5% | estimated ± 0.6 pp, medium confidence |
| DRACO | 82.3% | estimated ± 3.0 pp, medium confidence |
| HealthBench (raw) | 47.1% | measured |
| HealthBench (length-adjusted) | 53.2% | measured |
| HealthBench Professional | 60.8% | measured |
| HealthBench Professional (raw) | 59.5% | measured |
| HLE-Verified | 54.6% | estimated ± 0.7 pp, medium confidence |
| LABBench2 | 81.0% | estimated ± 1.3 pp, medium confidence |
| MMMU | 89.8% | estimated ± 0.9 pp, medium confidence |
| Agents' Last Exam | 56.4% | measured |
| OpenHarmony Bench | 58.8% | estimated ± 2.4 pp, high confidence |
| SEC-Bench Pro | 66.3% | measured |
| MMLU | 95.5% | estimated ± 1.1 pp, low confidence |
| MMMU-Pro w/ Python | 83.5% | estimated ± 1.6 pp, high confidence |
| RealWorldQA | 89.1% | estimated ± 1.2 pp, low confidence |
| OmniDocBench 1.5 | 92.0% | estimated ± 6.4 pp, low confidence |