O3 Mini (Medium): benchmark results
O3 Mini evaluated at the medium reasoning-effort setting. Provider: OpenAI. Released 2025-01-31. Access: API.
Aggregate Rating 1631 ± 12, rank #598 of 2168 rated models, from 102 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Swallow - Japanese MT-Bench - Math | 99.7 | Judge Score (normalized, %) | 99.3 |
| ZebraLogic | 88.9 | Puzzle Accuracy (%) | 98.4 |
| Swallow - English MT-Bench - Coding | 84.2 | Judge Score (normalized, %) | 97 |
| Swallow - Japanese MT-Bench - Extraction | 77.4 | Judge Score (normalized, %) | 97 |
| AidanBench | 3368 | Novel Answers | 96.4 |
| SEAL - Agentic Tool Use (Chat) | 62.42 | Score | 93.9 |
| Swallow - Post-trained English - IFBench | 67.7 | Instruction-Level Strict Accuracy (%) | 92.5 |
| Swallow - Post-trained Japanese - M-IFEval-Ja | 84.1 | Instruction-Level Strict Accuracy (%) | 92.5 |
| Swallow - English MT-Bench - Reasoning | 88.6 | Judge Score (normalized, %) | 91 |
| ProLLM - StackEval | 97.4 | Score (%) | 90.7 |
| Swallow - Japanese MT-Bench - Average | 78.2 | Judge Score (normalized, %) | 89.6 |
| Swallow - Japanese MT-Bench - Coding | 82.8 | Judge Score (normalized, %) | 89.6 |
Interactive version: theaggregate.ai/model?slug=o3-mini-medium · How It Works · Data refreshed daily, snapshot 2026-10-20.