OpenAI launched o3-mini—a model that’s not just powerful, but also more affordable and optimized for STEM tasks like coding, math, and science.
Why does this matter?
It runs faster and costs 63% less than its predecessor.
It makes 39% fewer major mistakes on complex real-world questions.
It produces clearer responses 24% faster than o1-mini.
Oh, and it gets better.
o3-mini is now available in ChatGPT (yes, even for free users). But if you’re on ChatGPT Plus, Team, or Pro, you get higher rate limits and more control over reasoning effort—letting you choose between low, medium, or high-depth answers.
And for devs? It’s hitting OpenAI’s API with flexible pricing, making it a competitive alternative to models like DeepSeek’s R1.
The big picture?
This launch comes at a pivotal time for OpenAI as they navigate competition, IP concerns, and one of the largest tech financing rounds in history.