GPT-6 Sol (Low): benchmark results
Provider: OpenAI. Released 2026-09-22. Access: API.
Aggregate Rating 1796 ± 15, rank #149 of 2178 rated models, from 72 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA GDP.pdf - Real Estate - Criterion Pass Rate | 78.79 | Criterion Pass Rate (%) | 96.7 |
| EnigmaForge - Fact F1 | 96.6 | World-fact recovery F1 (%) over 600 instances, all items | 92.6 |
| EnigmaForge - Intuition | 75 | Task success (%) on implicit-condition instances, where the | 92.6 |
| MLCR-AA - Accuracy | 90.56 | Accuracy (%) | 92 |
| AA GDP.pdf - Construction - Criterion Pass Rate | 75.48 | Criterion Pass Rate (%) | 91.4 |
| EnigmaForge | 74.2 | Task success (%) over 600 procedurally generated story insta | 88.9 |
| AA CritPt | 16.29 | Accuracy (%) | 83.8 |
| MLCR-AA | 25.56 | Overall Score (%) | 83.4 |
| AA Long Context Reasoning | 79.33 | Accuracy (%) | 81.7 |
| AA GDP.pdf | 24.2 | All Criteria Pass Rate (%) | 81.4 |
| AA Humanity's Last Exam | 34.94 | Accuracy (%) | 80.5 |
| AA MMMU-Pro | 80.17 | Accuracy (%) | 80.2 |
Interactive version: theaggregate.ai/model?slug=gpt-6-sol-low · How It Works · Data refreshed daily, snapshot 2026-11-08.