OpenAI · model
GPT-5.5 benchmark scores
As of 2026-10-07, GPT-5.5 (OpenAI) has measured scores on 54 benchmarks and estimated scores on 75 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 93.5% | measured |
| AA-HLE | 45.8% | measured |
| AA-SciCode | 55.8% | measured |
| AIME26 | 100.0% | estimated ± 0.6 pp, low confidence |
| Artificial Analysis Intelligence Index | 38.4% | measured |
| CritPt | 27.1% | measured |
| GDPval-AA | 42.7% | measured |
| GPQA Diamond | 93.6% | measured |
| HLE | 52.2% | measured |
| IFBench | 74.5% | estimated ± 6.9 pp, medium confidence |
| AA-LCR | 84.3% | measured |
| LiveCodeBench v6 | 89.0% | estimated ± 2.1 pp, high confidence |
| AA-Omniscience Accuracy | 58.0% | measured |
| SciCode | 47.3% | estimated ± 2.9 pp, medium confidence |
| Terminal-Bench 2.1 | 85.8% | estimated ± 6.9 pp, medium confidence |
| BrowseComp | 84.4% | measured |
| HLE w/ tools | 57.6% | estimated ± 4.6 pp, high confidence |
| IFEval | 93.7% | estimated ± 1.3 pp, high confidence |
| LongBench v2 | 61.5% | estimated ± 5.9 pp, low confidence |
| VITA-Bench | 46.3% | estimated ± 9.9 pp, low confidence |
| τ²-bench results | 98.0% | measured |
| AA Coding Index | 74.9% | measured |
| AA-MMMU-Pro | 79.9% | measured |
| APEX-Agents | 30.1% | estimated ± 0.6 pp, medium confidence |
| APEX-Agents-AA | 37.7% | measured |
| BioMysteryBench (human-difficult) | 48.7% | estimated ± 5.9 pp, low confidence |
| DeepSearchQA | 91.1% | estimated ± 4.5 pp, high confidence |
| SWE-bench Verified | 85.7% | estimated ± 0.8 pp, medium confidence |
| AutomationBench | 28.7% | estimated ± 5.8 pp, medium confidence |
| BFCL v4 | 76.8% | estimated ± 2.6 pp, medium confidence |
| CyberGym | 81.8% | measured |
| JobBench | 42.7% | measured |
| SWE-bench Pro | 58.6% | measured |
| τ³-bench results | 69.6% | estimated ± 1.8 pp, low confidence |
| WideResearch | 79.7% | estimated ± 1.2 pp, medium confidence |
| AA Agentic Index | 37.3% | measured |
| MMLU-Pro | 85.6% | estimated ± 2.5 pp, low confidence |
| FrontierMath v2 (Tier 4) | 35.4% | measured |
| FrontierMath v2 (Tiers 1-3) | 51.7% | measured |
| GPQA | 93.6% | measured |
| AA-IFBench | 75.9% | measured |
| Gert Labs | 72.9% | measured |
| AA-AnalystAgent | 50.0% | measured |
| AA EnterpriseOps-Gym | 52.4% | estimated ± 1.8 pp, low confidence |
| AA Harvey LAB | 91.4% | estimated ± 1.2 pp, medium confidence |
| ApprenticeBench | 20.0% | measured |
| ARC-AGI-1 | 95.0% | measured |
| ARC-AGI-2 | 85.0% | measured |
| CursorBench 3.1 | 59.2% | measured |
| CursorBench 3.2 | 58.4% | measured |
| Terminal-Bench 3.0 | 21.0% | estimated ± 6.9 pp, medium confidence |
| FrontierCode 1.1 Main | 43.0% | measured |
| FrontierSWE v2 | 12.6% | estimated ± 9.0 pp, medium confidence |
| OfficeQA Pro | 54.1% | measured |
| OSWorld-Verified | 78.7% | measured |
| Vals GPQA Diamond | 93.2% | measured |
| Vals LiveCodeBench | 85.3% | measured |
| Vals MMLU-Pro | 88.1% | measured |
| Vals SWE-bench | 82.6% | measured |
| Terminal-Bench 2.1 (Vals AI) | 76.4% | measured |
| VulcanBench v3 | 79.6% | estimated ± 5.1 pp, low confidence |
| AA AutomationBench | 51.5% | estimated ± 5.0 pp, medium confidence |
| GDP.pdf | 25.5% | estimated ± 3.8 pp, high confidence |
| AA ITBench | 45.8% | measured |
| AA Tau3 Banking | 47.1% | estimated ± 3.0 pp, low confidence |
| AA Terminal-Bench 2.1 | 87.1% | estimated ± 4.4 pp, high confidence |
| AA Terminal-Bench 4.0 | 32.2% | estimated ± 14.0 pp, low confidence |
| CursorBench 4.0 | 39.2% | estimated ± 1.6 pp, high confidence |
| CWE-bench v1 | 50.7% | estimated ± 6.9 pp, medium confidence |
| DeepSWE | 66.0% | estimated ± 3.4 pp, low confidence |
| HLE w/o tools | 41.4% | measured |
| OSWorld 2.0 | 13.0% | measured |
| PostTrainBench v1.1 | 27.2% | measured |
| ProgramBench | 0.0% | estimated ± 4.8 pp, low confidence |
| Toolathlon-Verified | 68.8% | estimated ± 1.0 pp, low confidence |
| Vibe Code Bench | 69.8% | measured |
| CharXiv | 84.5% | estimated ± 5.4 pp, medium confidence |
| CharXiv w/o tools | 85.7% | estimated ± 1.8 pp, medium confidence |
| ExploitBench | 20.0% | estimated ± 12.2 pp, low confidence |
| ExploitGym | 13.4% | measured |
| C-Eval | 93.3% | estimated ± 0.8 pp, low confidence |
| Claw-Eval | 61.0% | estimated ± 3.2 pp, high confidence |
| DeepPlanning | 47.8% | estimated ± 13.1 pp, low confidence |
| HMMT Nov 2025 | 93.2% | estimated ± 2.6 pp, low confidence |
| MathVision | 92.1% | estimated ± 3.6 pp, high confidence |
| MCP Atlas | 75.3% | measured |
| MCP-Tasks | 77.0% | estimated ± 0.6 pp, low confidence |
| MMAnswerBench | 89.9% | estimated ± 1.3 pp, low confidence |
| MMLU-Redux | 97.8% | estimated ± 1.1 pp, low confidence |
| MMMU-Pro | 81.2% | measured |
| NL2Repo | 47.7% | estimated ± 4.7 pp, high confidence |
| QwenClawBench | 62.6% | estimated ± 2.6 pp, high confidence |
| ScreenSpot Pro | 80.8% | estimated ± 10.0 pp, low confidence |
| SuperGPQA | 88.4% | estimated ± 7.5 pp, low confidence |
| Terminal-Bench 2.0 | 82.0% | measured |
| Toolathlon | 55.6% | measured |
| VideoMMMU | 87.8% | estimated ± 1.0 pp, high confidence |
| ERQA | 68.4% | estimated ± 3.9 pp, high confidence |
| HealthBench Hard | 36.1% | estimated ± 5.1 pp, low confidence |
| LiveCodeBench Pro | 82.6% | estimated ± 6.0 pp, low confidence |
| MedXpertQA (MM) | 77.1% | estimated ± 2.4 pp, high confidence |
| MedXpertQA (Text) | 66.3% | estimated ± 4.5 pp, medium confidence |
| MMLU-Pro (Arcee) | 88.1% | estimated ± 3.4 pp, low confidence |
| React Native Evals | 84.7% | measured |
| ResearchClawBench | 17.0% | measured |
| SWE-Rebench | 62.2% | estimated ± 6.5 pp, low confidence |
| ARC-AGI-3 | 0.4% | measured |
| FrontierMath (legacy) | 51.7% | measured |
| BioMysteryBench (human-solvable) | 89.5% | estimated ± 0.6 pp, medium confidence |
| DRACO | 79.1% | estimated ± 3.0 pp, medium confidence |
| FrontierCode 1.1 Extended | 55.1% | estimated ± 1.6 pp, low confidence |
| HealthBench (raw) | 54.3% | estimated ± 6.4 pp, low confidence |
| HealthBench (length-adjusted) | 53.8% | estimated ± 0.9 pp, medium confidence |
| HealthBench Professional | 59.5% | estimated ± 1.3 pp, medium confidence |
| HealthBench Professional (raw) | 64.3% | estimated ± 5.0 pp, medium confidence |
| HLE-Verified | 54.1% | estimated ± 0.7 pp, medium confidence |
| LABBench2 | 80.2% | estimated ± 1.3 pp, low confidence |
| MMMU | 87.2% | estimated ± 0.9 pp, medium confidence |
| LiveCodeBench | 100.0% | estimated ± 8.9 pp, low confidence |
| Agents' Last Exam | 24.2% | estimated ± 2.1 pp, high confidence |
| OpenHarmony Bench | 58.8% | estimated ± 2.4 pp, high confidence |
| SEC-Bench Pro | 45.8% | measured |
| SimpleVQA | 68.8% | estimated ± 8.2 pp, medium confidence |
| MMMLU | 90.9% | estimated ± 1.1 pp, low confidence |
| MMLU | 95.8% | estimated ± 1.1 pp, low confidence |
| MMMU-Pro w/ Python | 83.2% | measured |
| RealWorldQA | 87.6% | estimated ± 1.2 pp, medium confidence |
| Video-MME (with subtitle) | 89.1% | estimated ± 1.8 pp, medium confidence |
| OmniDocBench 1.5 | 91.7% | estimated ± 5.9 pp, low confidence |