O3 Mini (High): benchmark results
O3 Mini evaluated at the high reasoning-effort setting. Provider: OpenAI. Released 2025-01-31. Access: API.
Aggregate Rating 1606 ± 8, rank #692 of 2168 rated models, from 281 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| BBEH | 44.8 | Harmonic Mean (%) | 100 |
| COLLIE | 98.7 | Score (%) | 100 |
| Can LLMs Falsify? | 8.9 | Counterexample Rate (%) | 100 |
| OpenCompass Math - Reasoning Computation - Chinese (CompassBench 2504) | 65 | Score (%) | 100 |
| SEAL - Agentic Tool Use (Chat) | 63.45 | Score | 100 |
| SciCode | 34.4 | Subproblem Resolve Rate (%) | 100 |
| SuperCLUE General (March 2025) - Code Generation | 88.78 | Score | 100 |
| SuperCLUE General (March 2025) - Math Reasoning | 94.74 | Score | 100 |
| SuperCLUE General (March 2025) - Overall | 76.01 | Score | 100 |
| SuperCLUE General (March 2025) - Precise Instruction Following | 66.4 | Score | 100 |
| SuperCLUE-Project - Overall | 82.08 | Score | 100 |
| AidanBench | 4921 | Novel Answers | 98.2 |
Interactive version: theaggregate.ai/model?slug=o3-mini-high · How It Works · Data refreshed daily, snapshot 2026-10-20.