OpenAI · model
GPT-4.5 benchmark scores
As of 2026-10-29, GPT-4.5 (OpenAI) has measured scores on 4 benchmarks and estimated scores on 20 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 67.4% | estimated ± 6.3 pp, medium confidence |
| AA-HLE | 7.1% | estimated ± 5.0 pp, high confidence |
| CritPt | 0.4% | estimated ± 4.0 pp, high confidence |
| GPQA Diamond | 69.7% | estimated ± 1.2 pp, high confidence |
| HLE | 8.7% | estimated ± 6.9 pp, medium confidence |
| MMLU-Pro | 81.8% | estimated ± 4.0 pp, high confidence |
| Artificial Analysis Intelligence Index | 9.6% | measured |
| BBH | 96.6% | estimated ± 2.4 pp, low confidence |
| GPQA | 69.5% | measured |
| MMLU | 90.8% | measured |
| AA-Omniscience Accuracy | 30.0% | estimated ± 1.5 pp, high confidence |
| MMMLU | 82.8% | estimated ± 1.7 pp, high confidence |
| AA MMLU-Pro | 84.8% | estimated ± 1.0 pp, low confidence |
| ARC-AGI-1 | 71.7% | estimated ± 8.1 pp, low confidence |
| ARC-AGI-2 | 22.4% | estimated ± 14.9 pp, low confidence |
| Vals GPQA Diamond | 81.3% | estimated ± 5.3 pp, low confidence |
| Vals MMLU-Pro | 83.9% | estimated ± 2.1 pp, medium confidence |
| HLE w/o tools | 29.0% | estimated ± 4.6 pp, medium confidence |
| C-Eval | 88.4% | estimated ± 1.0 pp, medium confidence |
| MMLU-Redux | 90.1% | estimated ± 1.2 pp, medium confidence |
| SuperGPQA | 49.5% | estimated ± 8.2 pp, low confidence |
| HealthBench (length-adjusted) | 53.9% | estimated ± 3.3 pp, medium confidence |
| HealthBench Professional (raw) | 50.4% | estimated ± 8.6 pp, low confidence |
| SimpleQA | 62.5% | measured |