OpenAI Unveils o1 Pro, a New Reasoning Model That Tops Earlier Versions
Read the full article for context, quotes, and updates from the team.
OpenAI has released o1 Pro, a new reasoning model designed to deliver stronger performance on complex tasks than its o1-preview predecessor. The company says the model shows notable gains in areas that require multi-step thinking, including mathematics, coding, and other benchmark-heavy evaluations.
According to OpenAI, o1 Pro has achieved top results on several challenging tests, including a 93% score on the AIME 2024 math competition benchmark. The model is positioned as a major step forward in AI reasoning, reflecting continued progress in systems that can solve harder problems with greater accuracy and consistency.
The release is currently available to Pro and Team users, giving paying customers access to OpenAI’s latest reasoning-focused model. The launch underscores the company’s push to improve not just language generation, but the underlying problem-solving capabilities of its AI systems.
OpenAI has increasingly emphasized reasoning as a key frontier in artificial intelligence, especially for applications that depend on reliable analysis, code generation, and advanced mathematical performance. With o1 Pro, the company is signaling that it sees substantial room for improvement beyond earlier model versions, particularly in tasks where careful step-by-step reasoning matters most.
