OpenAI · model
GPT-6.1 Terra benchmark scores
As of 2026-10-29, GPT-6.1 Terra (OpenAI) has measured scores on 25 benchmarks and estimated scores on 92 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 96.1% | estimated ± 2.1 pp, medium confidence |
| AA-HLE | 46.2% | measured |
| AA-SciCode | 55.1% | measured |
| AIME26 | 98.1% | estimated ± 1.4 pp, medium confidence |
| CritPt | 28.0% | measured |
| GDPval-AA | 56.9% | estimated ± 2.9 pp, medium confidence |
| GPQA Diamond | 95.8% | estimated ± 1.0 pp, medium confidence |
| HLE | 49.8% | estimated ± 9.7 pp, medium confidence |
| AA-LCR | 82.7% | measured |
| LiveCodeBench v6 | 95.6% | estimated ± 4.9 pp, medium confidence |
| SciCode | 54.8% | estimated ± 3.1 pp, high confidence |
| Terminal-Bench 2.1 | 87.3% | estimated ± 5.2 pp, low confidence |
| BrowseComp | 91.2% | estimated ± 3.9 pp, medium confidence |
| HLE w/ tools | 59.7% | estimated ± 5.2 pp, medium confidence |
| LongBench v2 | 62.7% | estimated ± 5.8 pp, low confidence |
| AA Coding Index | 71.9% | estimated ± 6.1 pp, medium confidence |
| AA-MMMU-Pro | 84.7% | measured |
| BioMysteryBench (human-difficult) | 49.5% | estimated ± 6.8 pp, low confidence |
| SWE-bench Verified | 87.9% | estimated ± 6.4 pp, medium confidence |
| AutomationBench | 47.5% | estimated ± 11.5 pp, low confidence |
| JobBench | 69.4% | estimated ± 12.9 pp, low confidence |
| SWE-bench Pro | 67.2% | estimated ± 6.6 pp, medium confidence |
| AA Agentic Index | 50.2% | estimated ± 8.0 pp, low confidence |
| MMLU-Pro | 86.2% | estimated ± 4.1 pp, high confidence |
| Artificial Analysis Intelligence Index | 51.0% | measured |
| GPQA | 95.8% | estimated ± 1.0 pp, medium confidence |
| MMLU | 94.8% | estimated ± 1.0 pp, medium confidence |
| AA-Omniscience Accuracy | 47.6% | measured |
| MMMLU | 90.9% | estimated ± 2.6 pp, low confidence |
| MMMU | 89.2% | estimated ± 0.8 pp, medium confidence |
| AA MMLU-Pro | 88.3% | estimated ± 0.7 pp, high confidence |
| AA-AnalystAgent | 52.3% | estimated ± 2.8 pp, medium confidence |
| AA Tau3 Banking | 46.6% | estimated ± 5.1 pp, low confidence |
| AA Terminal-Bench 2.1 | 87.3% | estimated ± 4.8 pp, medium confidence |
| ApprenticeBench | 50.2% | estimated ± 12.5 pp, low confidence |
| ARC-AGI-1 | 93.4% | estimated ± 4.4 pp, low confidence |
| ARC-AGI-2 | 78.3% | estimated ± 14.6 pp, low confidence |
| CursorBench 3.1 | 56.6% | estimated ± 3.4 pp, medium confidence |
| CursorBench 3.2 | 61.8% | estimated ± 5.3 pp, medium confidence |
| FrontierCode 1.1 Main | 46.6% | estimated ± 6.6 pp, medium confidence |
| FrontierMath v2 (Tier 4) | 92.7% | measured |
| FrontierMath v2 (Tiers 1-3) | 90.9% | measured |
| FrontierSWE v2 | 29.4% | estimated ± 14.6 pp, low confidence |
| MazeBench | 16.0% | measured |
| OfficeQA Pro | 67.6% | estimated ± 6.8 pp, medium confidence |
| ParseBench | 73.6% | measured |
| Vals GPQA Diamond | 98.2% | estimated ± 1.5 pp, low confidence |
| Vals MMLU-Pro | 91.8% | estimated ± 1.5 pp, low confidence |
| Vals SWE-bench | 85.4% | estimated ± 7.5 pp, low confidence |
| Terminal-Bench 2.1 (Vals AI) | 81.0% | estimated ± 5.4 pp, low confidence |
| AA AutomationBench | 60.2% | measured |
| CWE-Bench-AA | 55.1% | estimated ± 8.4 pp, low confidence |
| DeepsecBench-AA | 22.6% | estimated ± 5.3 pp, medium confidence |
| GDP.pdf | 28.6% | measured |
| AA Terminal-Bench 4.0 | 55.6% | measured |
| CursorBench 4.0 | 39.2% | estimated ± 3.5 pp, high confidence |
| CWE-bench v1 | 62.3% | estimated ± 2.7 pp, medium confidence |
| DeepSWE | 65.5% | estimated ± 7.5 pp, low confidence |
| GraphWalks BFS 256K–1M | 71.9% | estimated ± 14.8 pp, low confidence |
| HLE w/o tools | 56.4% | estimated ± 4.6 pp, high confidence |
| OSWorld 2.0 | 53.1% | estimated ± 9.2 pp, low confidence |
| OSWorld 2.1 | 79.5% | estimated ± 2.8 pp, medium confidence |
| PostTrainBench v1.1 | 32.5% | estimated ± 7.4 pp, low confidence |
| SWE-bench Multimodal | 54.1% | estimated ± 7.2 pp, low confidence |
| Terminal-Bench 4.0 | 50.3% | measured |
| Toolathlon-Verified | 78.6% | estimated ± 1.3 pp, low confidence |
| OSWorld 2.1 Offline | 74.2% | measured |
| ARC-AGI-3 | 1.4% | estimated ± 7.7 pp, medium confidence |
| ARC-AGI-3 Provider Adapter | 10.7% | estimated ± 7.5 pp, medium confidence |
| Hermes Bench | 68.4% | estimated ± 3.4 pp, high confidence |
| Hermes Index | 58.9% | estimated ± 3.1 pp, high confidence |
| CharXiv | 93.4% | estimated ± 5.2 pp, low confidence |
| CharXiv w/o tools | 86.9% | estimated ± 1.7 pp, low confidence |
| ExploitBench | 97.1% | measured |
| ExploitGym | 37.0% | estimated ± 4.0 pp, high confidence |
| AI2D_TEST | 94.2% | estimated ± 2.0 pp, low confidence |
| C-Eval | 93.3% | estimated ± 1.5 pp, low confidence |
| CountBench | 100.0% | estimated ± 2.9 pp, low confidence |
| DynaMath | 85.5% | estimated ± 3.0 pp, low confidence |
| ERQA | 69.1% | estimated ± 2.9 pp, medium confidence |
| HMMT Feb 2026 | 100.0% | estimated ± 3.3 pp, medium confidence |
| HMMT Nov 2025 | 96.1% | estimated ± 1.5 pp, low confidence |
| MathVision | 95.3% | estimated ± 2.9 pp, medium confidence |
| MLVU (M-Avg) | 89.3% | estimated ± 2.2 pp, medium confidence |
| MMAnswerBench | 97.0% | estimated ± 2.4 pp, low confidence |
| MMLU-Redux | 96.9% | estimated ± 0.8 pp, low confidence |
| MMMU-Pro | 85.5% | measured |
| NL2Repo | 58.7% | estimated ± 7.4 pp, medium confidence |
| RealWorldQA | 88.9% | estimated ± 2.7 pp, medium confidence |
| ScreenSpot Pro | 79.5% | estimated ± 10.5 pp, low confidence |
| SimpleVQA | 75.1% | estimated ± 7.5 pp, low confidence |
| SuperGPQA | 95.1% | estimated ± 9.3 pp, low confidence |
| V* | 100.0% | estimated ± 7.5 pp, low confidence |
| Video-MME (w/o subtitle) | 88.2% | estimated ± 1.4 pp, low confidence |
| Video-MME (with subtitle) | 89.9% | estimated ± 3.0 pp, medium confidence |
| VideoMMMU | 99.8% | estimated ± 5.1 pp, low confidence |
| We-Math | 85.5% | estimated ± 7.4 pp, low confidence |
| HealthBench Hard | 31.4% | measured |
| MedXpertQA (MM) | 85.2% | estimated ± 2.4 pp, medium confidence |
| React Native Evals | 78.8% | estimated ± 4.7 pp, medium confidence |
| Finance Agent v2 | 62.3% | estimated ± 3.8 pp, medium confidence |
| BioMysteryBench (human-solvable) | 89.5% | estimated ± 0.6 pp, medium confidence |
| HealthBench (raw) | 51.1% | measured |
| HealthBench (length-adjusted) | 55.1% | measured |
| HealthBench Professional | 62.7% | measured |
| HealthBench Professional (raw) | 64.2% | measured |
| HLE-Verified | 54.3% | estimated ± 6.8 pp, low confidence |
| LABBench2 | 82.4% | estimated ± 2.4 pp, medium confidence |
| STEMBench | 2.6% | estimated ± 0.2 pp, high confidence |
| Agents' Last Exam | 60.0% | measured |
| OpenHarmony Bench | 58.9% | estimated ± 2.4 pp, high confidence |
| SEC-Bench Pro | 83.3% | measured |
| MMVU | 81.6% | estimated ± 2.7 pp, low confidence |
| MRCRv2 | 85.9% | estimated ± 11.3 pp, low confidence |
| MMLongBench-Doc | 61.5% | estimated ± 2.6 pp, low confidence |
| MMMU-Pro w/ Python | 87.0% | estimated ± 2.0 pp, medium confidence |
| OCRBench V2 | 74.4% | estimated ± 4.4 pp, low confidence |