---
title: "o3-mini: Reasoning with variable computational effort from OpenAI"
url: https://comunicagenia.com/en/blog/o3-mini-reasoning-with-variable-computational-effo/
date: 2026-10-09T19:15:53+00:00
modified: 2026-10-09T19:15:53+00:00
lang: en_US
---

# o3-mini: Reasoning with variable computational effort from OpenAI

**o3-mini** is the OpenAI model that revolutionizes the use of deliberative reasoning by allowing processing time to be adjusted at three levels: **low (Low), medium (Medium) and high (High)**.

## Dynamic Flexibility of Reasoning

This technical innovation allows users and developers to choose the exact balance between latency, token consumption and reflective depth:

### Low Level

Speed

Very quick responses to technical queries and everyday programming questions.

### Medium Level (Medium)

Balance

Default settings that maximize quality in complex code analysis.

### High Level

Maximum Analysis

Comprehensive dedication of reasoning tokens to critical engineering problems.

## Integration with Development Tools

o3-mini includes support for web searching, function calling, and output structuring, making it ideal as a stand-alone encoding agent.

Consult all the plans and models in our [definitive guide to OpenAI and ChatGPT](https://comunicagenia.com/en/openai-models-features-and-pricing/).