o3 Mini
ModelActiveby OpenAI · family “o3-mini”
OpenAI o3-mini is a cost-efficient language model optimized for STEM reasoning tasks, particularly excelling in science, mathematics, and coding. This model supports the `reasoning_effort` parameter, which can be set to... (description from the OpenRouter listing)
in: textin: fileout: textReasoningTool useStructured output
Epoch Capabilities Index
140.3 #108 of 268
90% CI 136.7 – 141.8
Released
31 Jan 2025
Context window
200K tokens
Max output
100K tokens
Input $ / 1M tokens
$1.10
Output $ / 1M tokens
$4.40
Cached input $ / 1M
$0.55
Each benchmark shown separately with its own source
Benchmark results (19)
| Benchmark | Domain | Score | vs best recorded | Setting | Run | Source |
|---|---|---|---|---|---|---|
| LMCA | agents | 19.0% | 28% | high | — | External ↗ |
| DTBench | reasoning | 68.8% | 70% | high | — | External ↗ |
| Mystery Game Puzzles | games | 7.0% ±2.6 | 8% | high | 27 Aug 2026 | Epoch ↗ |
| FrontierMath-Tiers-1-3-v2-Private | math | 18.6% ±2.3 | 20% | high | 11 Jun 2026 | Eval log ↗ |
| FrontierMath-Tier-4-v2-Private | math | 0.0% | 0% | high | 11 Jun 2026 | Eval log ↗ |
| Chess Puzzles | games | 17.0% ±3.8 | 24% | high | 8 Dec 2025 | Eval log ↗ |
| SimpleQA Verified | knowledge | 15.3% ±1.1 | 20% | high | 31 Aug 2026 | Eval log ↗ |
| FrontierMath-Tier-4-2025-07-01-Privatesuperseded | math | 4.2% ±2.9 | 9% | high | 1 Jul 2025 | Epoch ↗ |
| GSO-Bench | coding | 1.3% | 3% | low | — | External ↗ |
| ARC-AGI-2 | reasoning | 3.0% | 3% | high | — | External ↗ |
| FrontierMath-2025-02-28-Privatesuperseded | math | 12.4% ±1.9 | 24% | high | 16 Nov 2025 | Epoch ↗ |
| Lech Mazur Writing | other | 61.7% | 72% | high | — | External ↗ |
| WeirdML | coding | 43.7% ±0.0 | 47% | high | — | External ↗ |
| Aider polyglot | coding | 60.4% | 69% | high | — | External ↗ |
| OTIS Mock AIME 2024-2025 | math | 76.9% ±5.2 | 77% | high | 27 Feb 2025 | Eval log ↗ |
| SimpleBench | reasoning | 22.8% | 28% | high | — | External ↗ |
| GPQA diamond | science | 77.0% ±2.6 | 80% | high | 13 Feb 2025 | Eval log ↗ |
| MATH level 5 | math | 96.5% ±0.4 | 98% | high | 13 Feb 2025 | Epoch ↗ |
| ARC-AGI | reasoning | 34.5% | 35% | high | — | External ↗ |
Source: Epoch AI Benchmarking Hub (CC BY 4.0). “External” rows are leaderboard results Epoch collects from third parties. Best reported setting per benchmark is shown.
API list price over time
Price history
$ per 1M tokens
13 recorded prices on 1 Oct 2025 (OpenRouter listing + Internet Archive snapshots); re-read on every data refresh, most recently 8h ago. Steps show when the price changed.
Events
- BenchmarkEpoch AI evaluates o3 MiniEpoch AI Benchmarking Hub
- BenchmarkEpoch AI evaluates o3 MiniEpoch AI Benchmarking Hub