OpenAI has officially released o3-mini, its latest reasoning-focused AI model that represents a significant advancement in making powerful AI more accessible and cost-effective.
First previewed in December 2024 and launched in January 2025, o3-mini is specifically optimized for STEM applications, demonstrating particular strength in science, mathematics, and coding tasks. The model employs a chain-of-thought technique that allows it to "think" about responses before answering, providing step-by-step analysis of problems that leads to more accurate solutions.
Unlike traditional language models that rely on pattern recognition, o3-mini incorporates "simulated reasoning," enabling it to pause and reflect on internal thought processes before responding. This approach has yielded impressive results, with o3-mini making 39% fewer major mistakes on complex real-world questions compared to its predecessor, o1-mini, while delivering responses 24% faster.
The model offers three reasoning effort options—low, medium, and high—allowing users to optimize for specific use cases that prioritize either processing power for complex challenges or speed for time-sensitive applications. In benchmark tests, o3-mini (high) has demonstrated exceptional performance, achieving 87.3% accuracy on the AIME mathematical reasoning test and outperforming both o1-mini and the full o1 model on several key metrics.
Pricing is another significant advantage, with o3-mini costing 63% less than o1-mini at $0.55 per million cached input tokens and $4.40 per million output tokens. This competitive pricing makes advanced AI reasoning more accessible to developers and businesses of all sizes.
The model is available to all ChatGPT users, with free-tier users having limited access and paid subscribers enjoying higher usage limits. ChatGPT Plus and Team users can use o3-mini for up to 150 messages per day, while Pro subscribers have unlimited access. Developers can also integrate o3-mini through OpenAI's API, with the ability to customize reasoning effort based on their specific needs.
As of June 2025, o3-mini has been joined by newer models in OpenAI's lineup, including the full o3 model released in April and the more powerful o3-pro launched earlier this month. However, o3-mini remains a valuable option for those seeking a balance of performance, speed, and cost efficiency in AI reasoning capabilities.