O4 Mini (2025-04-16) (High): benchmark results
O4 Mini (2025-04-16) evaluated at the high reasoning-effort setting. Provider: OpenAI. Released 2025-04-16. Access: API.
Aggregate Rating 1639 ± 10, rank #562 of 2168 rated models, from 150 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Defects4J | 53.8 | Defects4J Plausible @1 (imported) | 100 |
| OpenCompass Agent - Hard Multi-Turn (CompassBench 2510) | 58 | Score (%) | 100 |
| OpenCompass LLM - Code - English (CompassBench 2507) | 51.4 | Score (%) | 100 |
| OpenCompass Language - Dialogue - English (CompassBench 2510) | 84.3 | Score (%) | 100 |
| RepairBench | 50.3 | Plausible@1 (%) | 100 |
| OpenCompass Code - Competition - Chinese (CompassBench 2510) | 69.6 | Score (%) | 98.1 |
| MATH Level 5 | 97.83 | Accuracy (%) | 97.2 |
| Arena-Hard v2 | 79.1 | Win Rate (%) | 96.3 |
| Arena-Hard v2 (GPT-4.1 Judge) | 81.7 | Win Rate (%) | 96.3 |
| OpenCompass Code - Execution - English (CompassBench 2507) | 31.8 | Score (%) | 96.3 |
| OpenCompass LLM - Code (CompassBench 2507) | 48.1 | Score (%) | 96.3 |
| OpenCompass Code - Competition - English (CompassBench 2510) | 73.9 | Score (%) | 96.2 |
Interactive version: theaggregate.ai/model?slug=o4-mini-2025-04-16-high · How It Works · Data refreshed daily, snapshot 2026-10-20.