OpenAI · model
GPT-5.4 Pro benchmark scores
As of 2026-10-07, GPT-5.4 Pro (OpenAI) has measured scores on 9 benchmarks and estimated scores on 50 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 93.4% | estimated ± 2.7 pp, high confidence |
| AA-HLE | 51.0% | estimated ± 4.6 pp, high confidence |
| AIME26 | 100.0% | estimated ± 0.6 pp, low confidence |
| Artificial Analysis Intelligence Index | 49.4% | estimated ± 5.1 pp, medium confidence |
| CritPt | 30.0% | measured |
| GDPval-AA | 51.1% | estimated ± 8.4 pp, medium confidence |
| GPQA Diamond | 93.3% | estimated ± 1.4 pp, high confidence |
| HLE | 58.7% | measured |
| AA-Omniscience Accuracy | 55.9% | estimated ± 6.2 pp, medium confidence |
| BrowseComp | 89.3% | measured |
| HLE w/ tools | 59.0% | estimated ± 4.6 pp, high confidence |
| VITA-Bench | 46.2% | estimated ± 9.9 pp, low confidence |
| APEX-Agents-AA | 39.9% | estimated ± 4.6 pp, high confidence |
| BioMysteryBench (human-difficult) | 49.7% | estimated ± 6.8 pp, low confidence |
| DeepSearchQA | 91.6% | estimated ± 8.1 pp, low confidence |
| AutomationBench | 37.5% | estimated ± 10.3 pp, low confidence |
| CyberGym | 82.4% | estimated ± 4.6 pp, high confidence |
| JobBench | 50.8% | estimated ± 10.1 pp, low confidence |
| WideResearch | 81.7% | estimated ± 4.6 pp, high confidence |
| AA Agentic Index | 47.6% | estimated ± 8.2 pp, medium confidence |
| MMLU-Pro | 85.7% | estimated ± 2.5 pp, low confidence |
| FrontierMath v2 (Tier 4) | 37.5% | measured |
| FrontierMath v2 (Tiers 1-3) | 50.0% | measured |
| GPQA | 93.7% | estimated ± 1.6 pp, high confidence |
| Gert Labs | 74.7% | estimated ± 7.4 pp, low confidence |
| AA-AnalystAgent | 41.8% | estimated ± 12.5 pp, low confidence |
| AA EnterpriseOps-Gym | 44.1% | estimated ± 6.9 pp, low confidence |
| AA Harvey LAB | 92.8% | estimated ± 9.8 pp, low confidence |
| ARC-AGI-1 | 94.5% | measured |
| ARC-AGI-2 | 83.3% | measured |
| OSWorld-Verified | 84.3% | estimated ± 5.4 pp, low confidence |
| Vals GPQA Diamond | 93.6% | estimated ± 2.2 pp, high confidence |
| Vals MMLU-Pro | 89.2% | estimated ± 1.8 pp, high confidence |
| AA AutomationBench | 53.9% | estimated ± 5.7 pp, low confidence |
| GDP.pdf | 15.3% | estimated ± 6.6 pp, low confidence |
| AA Tau3 Banking | 37.4% | estimated ± 10.2 pp, low confidence |
| HLE w/o tools | 42.7% | measured |
| OSWorld 2.0 | 67.8% | estimated ± 13.8 pp, low confidence |
| Toolathlon-Verified | 73.7% | estimated ± 7.7 pp, medium confidence |
| ExploitGym | 29.5% | estimated ± 6.1 pp, low confidence |
| C-Eval | 93.3% | estimated ± 0.8 pp, low confidence |
| Claw-Eval | 81.7% | estimated ± 7.9 pp, low confidence |
| HMMT Nov 2025 | 93.2% | estimated ± 2.6 pp, low confidence |
| MCP Atlas | 77.8% | estimated ± 10.0 pp, low confidence |
| MMAnswerBench | 90.4% | estimated ± 1.3 pp, low confidence |
| MMLU-Redux | 97.0% | estimated ± 2.1 pp, low confidence |
| SuperGPQA | 85.9% | estimated ± 9.1 pp, low confidence |
| Toolathlon | 56.4% | estimated ± 3.5 pp, high confidence |
| HealthBench Hard | 34.2% | estimated ± 9.0 pp, low confidence |
| MedXpertQA (Text) | 60.2% | estimated ± 8.9 pp, low confidence |
| MMLU-Pro (Arcee) | 100.0% | estimated ± 4.0 pp, low confidence |
| ResearchClawBench | 19.5% | estimated ± 2.2 pp, low confidence |
| ARC-AGI-3 | 5.6% | estimated ± 11.5 pp, low confidence |
| FrontierMath (legacy) | 50.0% | measured |
| BioMysteryBench (human-solvable) | 89.5% | estimated ± 0.6 pp, medium confidence |
| HealthBench (length-adjusted) | 56.8% | estimated ± 2.5 pp, low confidence |
| HealthBench Professional | 56.4% | estimated ± 4.6 pp, high confidence |
| HLE-Verified | 46.4% | estimated ± 0.7 pp, low confidence |
| LABBench2 | 79.3% | estimated ± 1.3 pp, low confidence |