O4 Mini (High): benchmark results
O4 Mini evaluated at the high reasoning-effort setting. Provider: OpenAI. Released 2025-04-16. Access: API.
Aggregate Rating 1658 ± 6, rank #496 of 2168 rated models, from 320 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| ConstraintBench - Order Fulfillment | 90 | Feasibility rate (%): share of the domain's 20 tasks whose r | 100 |
| HLCE | 15.84 | Pass@1 (%) | 100 |
| LLM2014 Code 2025-09 - Python | 9.17 | Score | 100 |
| LiveCodeBench | 87.3 | Pass@1 avg (%) | 100 |
| OpenCompass Code - Competition - Chinese (CompassBench 2504) | 83 | Score (%) | 100 |
| OpenCompass Code - Execution - English (CompassBench 2504) | 32.4 | Score (%) | 100 |
| OpenCompass LLM - Code - Chinese (CompassBench 2504) | 65.1 | Score (%) | 100 |
| OpenCompass LLM - Math (CompassBench 2504) | 68.8 | Score (%) | 100 |
| OpenCompass LLM - Math - English (CompassBench 2504) | 75.5 | Score (%) | 100 |
| OpenCompass Math - Competition - English (CompassBench 2504) | 90 | Score (%) | 100 |
| OpenCompass Math - Reasoning Computation - English (CompassBench 2504) | 80.6 | Score (%) | 100 |
| SuperCLUE General (May 2025) - Agent | 76.01 | Score | 100 |
Interactive version: theaggregate.ai/model?slug=o4-mini-high · How It Works · Data refreshed daily, snapshot 2026-10-20.