O3 (High): benchmark results
O3 evaluated at the high reasoning-effort setting. Provider: OpenAI. Released 2025-04-16. Access: API.
Aggregate Rating 1699 ± 8, rank #359 of 2168 rated models, from 204 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| CharXiv-D | 95 | Score (%) | 100 |
| CharXiv-R | 78.6 | Validation chart-visible reasoning accuracy (%) | 100 |
| Ducky Bench (Saxo Frog) | 1449 | ELO | 100 |
| EstGraph - Node Count, BA Large (MH Walk) | 13.08 | Median relative error of the node-count estimate (%; 100 syn | 100 |
| EstGraph - Node Count, BA Large (Simple Walk) | 25.47 | Median relative error of the node-count estimate (%; 100 syn | 100 |
| EstGraph - Node Count, BA Medium (MH Walk) | 9.63 | Median relative error of the node-count estimate (%; 100 syn | 100 |
| EstGraph - Node Count, BA Small (MH Walk) | 17.6 | Median relative error of the node-count estimate (%; 100 syn | 100 |
| EstGraph - Node Count, BA Small (Simple Walk) | 16.01 | Median relative error of the node-count estimate (%; 100 syn | 100 |
| EstGraph - Node Count, ER Large (MH Walk) | 3.41 | Median relative error of the node-count estimate (%; 100 syn | 100 |
| EstGraph - Node Count, ER Large (Simple Walk) | 5.57 | Median relative error of the node-count estimate (%; 100 syn | 100 |
| EstGraph - Node Count, ER Small (MH Walk) | 7.59 | Median relative error of the node-count estimate (%; 100 syn | 100 |
| EstGraph - Node Count, ER Small (Simple Walk) | 9.59 | Median relative error of the node-count estimate (%; 100 syn | 100 |
Interactive version: theaggregate.ai/model?slug=o3-high · How It Works · Data refreshed daily, snapshot 2026-10-20.