OpenAI responds to the DeepSeek buzz by launching its latest o3-mini reasoning model for all users
Posted by:
Sarah Collins
OpenAI has officially rolled out its latest o3-mini AI model to users of ChatGPT, including those on the free tier, following a prior teaser. The new model boasts enhanced reasoning abilities, particularly in areas such as math, coding, and science. It is said to outperform the previous o1-mini model with a 24% faster response time and is preferred over o1-mini responses 56% of the time, with 39% fewer errors, according to external testers. The model displays its reasoning process along with responses and can be paired with web searches, albeit in early stages. This release comes amidst growing competition, notably from China’s DeepSeek AI models that have garnered significant user interest for speed and accuracy. The o3-mini is available for use in ChatGPT with varying access levels for free, Plus, Team, and Pro users, along with a slightly slower but more advanced o3-mini-high model for paying users. OpenAI emphasizes the safety measures taken in the development of o3-mini, surpassing the GPT-4o model in detecting unsafe use and potential security breaches.
Recent Posts1
