GPT-6 (Low): benchmark results
Provider: OpenAI. Released 2026-09-03. Access: API.
Unified ELO 1753 ± 1, rank #26 of 1761 rated models, from 22 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA Omniscience - Software Engineering (SWE) | 89.1 | Accuracy (%) | 98.5 |
| AA Omniscience | 40.55 | Score | 98.4 |
| AA Omniscience - Health | 52.6 | Accuracy (%) | 98.4 |
| AA-Omniscience Accuracy | 59.53 | Accuracy (%) | 97.9 |
| AA Humanity's Last Exam | 49.21 | Accuracy (%) | 97.7 |
| AA Omniscience - Business | 49.8 | Accuracy (%) | 97.5 |
| AA Omniscience - Humanities & Social Sciences | 56.6 | Accuracy (%) | 97.1 |
| Artificial Analysis Intelligence Index | 49.32 | Intelligence Index | 96.8 |
| AA GPQA Diamond | 93.13 | Accuracy (%) | 96.1 |
| AA MMMU-Pro | 84.62 | Accuracy (%) | 96.1 |
| AA Omniscience - Law | 57.6 | Accuracy (%) | 96 |
| AA CritPt | 26.29 | Accuracy (%) | 95.8 |
Interactive version: theaggregate.ai/model?slug=gpt-6-low · How It Works · Data refreshed daily, snapshot 2026-09-05.