OpenAI Launches o3-Pro, Its Most Advanced AI Reasoning Model
OpenAI introduces o3-pro, a powerful new step-by-step reasoning model that outperforms competitors in math and science benchmarks. The update is rolling out immediately to premium subscribers and developers.
OpenAI announces the launch of o3-pro, its most advanced reasoning artificial intelligence model to date. This new model is an upgraded version of the o3 system introduced earlier this year and relies on step-by-step problem solving to deliver highly reliable results in complex fields like physics, mathematics, and computer programming.
The o3-pro model is available immediately for ChatGPT Pro and Team users, replacing the older o1-pro model, while Enterprise and Edu users gain access the following week. Developers access the model through the OpenAI API, which costs 20 dollars per million input tokens and 80 dollars per million output tokens. Expert reviewers consistently prefer o3-pro over the standard o3 in categories like science, programming, business, and writing assistance.
Equipped with web browsing, file analysis, visual reasoning, and Python execution tools, o3-pro achieves impressive results on major industry benchmarks. It outperforms Google's Gemini 2.5 Pro on the AIME 2024 mathematical assessment and surpasses Anthropic's Claude 4 Opus on the GPQA Diamond test for PhD-level science knowledge.