O1 Preview: benchmark results
OpenAI's September 2024 preview release of the o1 reasoning model. Provider: OpenAI. Released 2024-09-12. Access: API.
Aggregate Rating 1702 ± 9, rank #193 of 1656 rated models, from 117 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| CSimpleQA | 67.9 | Score (%) | 100 |
| Divergent Thinking | 4.79 | Divergence Score | 100 |
| FullStackBench (2024-12 bilingual) | 65.62 | Pass@1 (%), 3,374 problems: 1,687 English and 1,687 Chinese; | 100 |
| MixEval | 72 | Score | 100 |
| SEAL - Japanese | 1118 | Score | 100 |
| SuperCLUE General (October 2024) - Hard | 64.89 | Score | 100 |
| SuperCLUE General (October 2024) - Overall | 75.85 | Score | 100 |
| SuperCLUE General (October 2024) - Science | 86.07 | Score | 100 |
| WebApp1K-React | 0.95 | pass@1 | 100 |
| Enkrypt AI - Jailbreak Risk | 0.86 | Risk Score | 98.3 |
| ChemBench | 64 | Overall Score (%) | 97.3 |
| SuperCLUE General (December 2024) - Hard | 63.6 | Score | 97.3 |
Interactive version: theaggregate.ai/model?slug=o1-preview · How It Works · Data refreshed daily, snapshot 2026-10-20.