O3 (2025-04-16) (Medium): benchmark results
O3 (2025-04-16) evaluated at the medium reasoning-effort setting. Provider: OpenAI. Released 2025-04-16. Access: API.
Aggregate Rating 1702 ± 15, rank #353 of 2168 rated models, from 51 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Fiction.LiveBench | 100 | Accuracy (%) | 100 |
| Swallow - English MT-Bench - Humanities | 82.9 | Judge Score (normalized, %) | 97 |
| Swallow - English MT-Bench - Math | 99.6 | Judge Score (normalized, %) | 97 |
| Swallow - English MT-Bench - Reasoning | 90.3 | Judge Score (normalized, %) | 97 |
| Swallow - Post-trained Japanese - JamC-QA | 85.1 | Accuracy (%) | 97 |
| UGI - Writing | 66.83 | Writing Score | 95.8 |
| Swallow - English MT-Bench - Average | 84.9 | Judge Score (normalized, %) | 95.5 |
| Swallow - English MT-Bench - Roleplay | 82.8 | Judge Score (normalized, %) | 95.5 |
| Swallow - English MT-Bench - STEM | 82.5 | Judge Score (normalized, %) | 95.5 |
| Swallow - English MT-Bench - Writing | 79.7 | Judge Score (normalized, %) | 95.5 |
| Swallow - Japanese MT-Bench - Average | 81.8 | Judge Score (normalized, %) | 95.5 |
| Swallow - Japanese MT-Bench - Roleplay | 79.6 | Judge Score (normalized, %) | 95.5 |
Interactive version: theaggregate.ai/model?slug=o3-2025-04-16-medium · How It Works · Data refreshed daily, snapshot 2026-10-20.