o3 mini
by OpenAI
+
+
+
+
About
OpenAI o3 mini, released in January 2025, brought the reasoning capabilities of the o3 training approach to a compact, cost-efficient model targeting STEM problem-solving. It was notable for being the first model to score competitively with full reasoning models while remaining practical for production API usage — offering developers a path to capable mathematical and scientific reasoning without the cost ceiling of full o3.
+
+
+
+
Pricing Range
Input (per 1M)$1.10 -$1.10
Output (per 1M)$4.40 -$4.40
Providers1
+
+
+
+
Timeline
ReleasedJan 31, 2025
+
+
+
+
License & Family
License
Proprietary
Performance Overview
Performance metrics and category breakdown
Overall Performance
1 benchmarks
Average Score
1.5%
Best Score
1.5%
High Performers (80%+)
0Performance Metrics
Max Context Window
300.0KTop Categories
Agents
1.5%
+
+
+
+
All Benchmark Results for o3 mini
Complete list of benchmark scores with detailed information
| BrowseComp | Agents | 1.50 | 1.5% | Unverified |