← LLM Models#12
o3-mini
OpenAI
Cost-efficient reasoning. Optimized for STEM - outperforms o1 on math.
Official page →📖Context128K tokens
🚀Speed150 tok/s
💵Input price$1.10/1M
💸Output price$4.40/1M
Benchmarks
Gray bar = dataset averageMMLU57 academic subjects
86.9%avg 81.8
GPQA DiamondPhD-level science questions
67.4%avg 50.4
HumanEvalPython code generation
91.7%avg 79.9
SWE-Bench VerifiedReal GitHub engineering tasks
49.3%avg 32.4
Composite ScoreScore = MMLU×20% + GPQA×30% + HumanEval×25% + SWE-Bench×25%72.9
More from OpenAI
Details
- Provider
- OpenAI
- Parameters
- -
- Context
- 128K tokens
- Speed
- 150 tok/s
- Open Source
- No
- License
- Proprietary
- Released
- Jan 2025