OpenAI · model
GPT-6 Astra (high) benchmark scores
As of 2026-10-07, GPT-6 Astra (high) (OpenAI) has measured scores on 5 benchmarks and estimated scores on 7 more.
| Benchmark | Score | Source |
|---|---|---|
| Terminal-Bench 2.1 | 89.9% | measured |
| Terminal-Bench 4.0 | 47.8% | estimated ± 11.6 pp, low confidence |
| GPQA Diamond | 94.9% | measured |
| CritPt | 25.9% | estimated ± 6.8 pp, medium confidence |
| GDP.pdf | 31.0% | measured |
| τ³-Bench Banking | 47.4% | estimated ± 5.0 pp, medium confidence |
| MMMU-Pro | 86.0% | measured |
| AA-AnalystAgent | 53.4% | estimated ± 7.8 pp, medium confidence |
| Harvey LAB | 93.9% | estimated ± 3.2 pp, high confidence |
| Humanity's Last Exam | 50.4% | estimated ± 4.6 pp, high confidence |
| AutomationBench | 66.6% | measured |
| EnterpriseOps-Gym | 42.3% | estimated ± 6.2 pp, low confidence |