OpenAI
o3-mini
o3-mini is OpenAI's most recent small reasoning model, providing high intelligence at the same cost and latency targets of o1-mini.
Introducing o3 and o4-miniModel Snapshot
TypeLanguage
ReleaseDec 20, 2024
Context200,000
Tagsfile-input, reasoning, tool-use, implicit-caching
Overall
Rank #185 across 526
25.9
Coding
Rank #231 across 436
17.9
Blended Price
Rank #249 across 357
$1.93/M
Strength Profile
Percentile score by analysis domain.
* Cost is inverted: lower input, output, and blended prices rank higher.
Benchmark Percentiles
Higher bars mean stronger relative placement.
Model Details
Runtime
Output Speed203.3 tok/s
First Token5.76s
Max Output100,000
Pricing
Blended$1.93/M
Input$1.10/M
All Benchmarks
| Metric | Domain | Value | Rank |
|---|---|---|---|
| Artificial Analysis Intelligence Index | overall | 25.9 | #185 |
| Artificial Analysis Coding Index | coding | 17.9 | #231 |
| MMLU-Pro | reasoning | 79.1% | #129 |
| GPQA | reasoning | 74.8% | #179 |
| Humanity's Last Exam |