OpenAI has released o3-pro, a new AI model they claim is their most advanced to date. This model is designed to enhance reasoning capabilities, making it more reliable in fields like science, mathematics, and coding.
O3-pro is an upgraded version of OpenAI's o3 model. It is now available for ChatGPT Pro and Team users, replacing the o1-pro model. Enterprise and Edu users will gain access the following week. The model is also live in OpenAI's developer API.
The new model is priced at $20 per million input tokens and $80 per million output tokens in the API. O3-pro has access to tools that allow it to search the web, analyze files, and use Python. However, temporary chats in ChatGPT are disabled due to a technical issue, and it cannot generate images.
In internal testing, o3-pro has shown impressive results. It outperforms Google's Gemini 2.5 Pro on the AIME 2024 math skills test. It also surpasses Anthropic's Claude 4 Opus on the GPQA Diamond test, which assesses PhD-level science knowledge.