OpenAI launches o3-pro AI model, offering increased reliability and tool use for enterprises — while sacrificing speed
1 min read
Summary
OpenAI’s new o3-pro model has been designed to think longer and more reliably, offering developers access to additional software tool integrations and a higher level of detail and accuracy, but it will be slower than the models they are likely used to.
The model is available to developers via the OpenAI API and for Pro and Team users of ChatGTP, replacing o1-pro in the model picker for paying ChatGPT users, but at a price, with o3-pro costing 20perinputand80 for output, compared to o3 itself, which is now down to 2and8.
OpenAI has said that o3-pro is better at following instructions, more effective and more comprehensive than o3, which was already impressive.
This latest development shows the direction of travel for AI, with reasoning models the new battleground for model providers, with Google, Anthropic, xAI and others all developing similar tools.