o3-mini (high)
Intelligence
25.2
#205/523
Coding
17.3
#236/429
Math
—
Speed
150 tok/s
TTFT: 24.44s
Pricing
$1.10 / $4.40
per 1M tokens (in/out)
A smaller variant of O3, expected to offer enhanced multimodal capabilities, improved reasoning, and more efficient resource utilization compared to previous models while maintaining strong performance on core tasks.
o3-mini (high) is OpenAI’s model designed for efficient processing of text data. It operates at a speed of 150.362 tokens per second and is priced at $1.1 per million input tokens, making it suitable for various professional applications.
When to Use o3-mini (high)
✓ Best For
- Text generation and completion tasks.
- Basic coding assistance.
- Data analysis and summarization.
✗ Not Ideal For
- Complex mathematical problem solving.
- High-speed real-time applications.
How o3-mini (high) Compares
Intelligence Index · Higher is better
GoogleChina MobileOpenAIxAIByteDance Seed
Benchmark Profile
Coding Index
NaverAlibabaOpenAIInclusionAI
Output Speed · tok/s
GoogleStepFunOpenAICohere
Intelligence · Coding · Math
Intelligence
Coding
Math
All Benchmark Scores (13)
| Benchmark | Score | |
|---|---|---|
| Intelligence Index | 25.2 | |
| Coding Index | 17.3 | |
| MMLU-Pro | 802% | |
| GPQA | 773% | |
| LiveCodeBench | 734% | |
| HLE | 123% | |
| SciCode | 39.8% | |
| IFBench | 67.1% | |
| LCR | 39.3% | |
| TerminalBench Hard | 6.1% | |
| Tau2 | 31.3% | |
| AIME | 86% | |
| MATH 500 | 98.5% |
Data: Artificial Analysis · Updated: March 28, 2026
Frequently Asked Questions (15)
When was o3-mini (high) released?
o3-mini (high) was released on January 31, 2025.
Who created o3-mini (high)?
o3-mini (high) was created by OpenAI.
How intelligent is o3-mini (high)?
o3-mini (high) scores 25 on the Artificial Analysis Intelligence Index, placing it below average among other reasoning models in a similar price tier (median: 31).
How fast is o3-mini (high)?
o3-mini (high) generates output at 125.1 tokens per second (based on OpenAI's API), which is well above average compared to other reasoning models in a similar price tier (median: 65.5 t/s).
What is the latency of o3-mini (high)?
o3-mini (high) has a time to first token (TTFT) of 41.14s (based on OpenAI's API), which is at the higher end compared to other reasoning models in a similar price tier (median: 2.66s).
How much does o3-mini (high) cost?
o3-mini (high) costs $1.10 per 1M input tokens (better than average, median: $1.35) and $4.40 per 1M output tokens (very competitive, median: $8.40), based on OpenAI's API.
What is o3-mini (high) API pricing?
o3-mini (high) costs $1.10 per 1M input tokens and $4.40 per 1M output tokens (based on OpenAI's API). For a blended rate (3:1 input to output ratio), this is $1.93 per 1M tokens. Pricing may vary by provider.
How verbose is o3-mini (high)?
When evaluated on the Intelligence Index, o3-mini (high) generated 61M output tokens, which is at the higher end compared to other reasoning models in a similar price tier (median: 13M).
Is o3-mini (high) a reasoning model?
Yes, o3-mini (high) is a reasoning model. It uses extended thinking or chain-of-thought reasoning to work through complex problems before providing an answer.
What input modalities does o3-mini (high) support?
o3-mini (high) supports text input.
What output modalities does o3-mini (high) support?
o3-mini (high) supports text output.
Can o3-mini (high) process images?
No, o3-mini (high) does not support image input. It can only process text.
Is o3-mini (high) multimodal?
No, o3-mini (high) is not multimodal. It only supports text input.
What is the context window of o3-mini (high)?
o3-mini (high) has a context window of 200k tokens. This determines how much text and conversation history the model can process in a single request.
Is o3-mini (high) open source?
No, o3-mini (high) is proprietary. The model weights are not publicly available.
