GPT-6 (Non-reasoning): benchmark results
Provider: OpenAI. Released 2026-09-03. Access: API.
Unified ELO 1729 ± 1, rank #53 of 1761 rated models, from 21 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Artificial Analysis Intelligence Index | 47.85 | Intelligence Index | 96.2 |
| AA Omniscience - Health | 48.9 | Accuracy (%) | 95.3 |
| AA Omniscience - Software Engineering (SWE) | 82.9 | Accuracy (%) | 95.2 |
| AA-Omniscience Accuracy | 55.6 | Accuracy (%) | 95.1 |
| AA Omniscience - Business | 46.8 | Accuracy (%) | 95 |
| AA Omniscience | 26.55 | Score | 94.9 |
| AA Omniscience - Humanities & Social Sciences | 54 | Accuracy (%) | 94.6 |
| AA Omniscience - Law | 51.6 | Accuracy (%) | 92.6 |
| AA CritPt | 19.71 | Accuracy (%) | 92.2 |
| AA Omniscience - Science, Engineering & Mathematics | 49.4 | Accuracy (%) | 91.4 |
| AA MMMU-Pro | 82.02 | Accuracy (%) | 91.3 |
| AA GDPval | 1546.73 | ELO | 89.9 |
Interactive version: theaggregate.ai/model?slug=gpt-6-non-reasoning · How It Works · Data refreshed daily, snapshot 2026-09-05.