ChatGPT maker OpenAI announced a new language learning model called o1 codenamed Strawberry that reasons through problems like a PhD student.
OpenAI claims its new model can reason through complex tasks and solve harder problems in coding, math, physics, biology, and chemistry than its previous models.
The o1 model is part of a family of models, with the first releases being o1-preview and o1-mini, a smaller, more efficient variant designed specifically for code generation.
ChatGPT Plus and Team users will be able to access o1 models in ChatGPT starting today while ChatGPT Enterprise and Edu users will get access to both models beginning next week.
The new model differs from previous versions by evaluating multiple solution paths before responding to a query, which results in slightly slower response times.
According to tests done by OpenAI, the o1 model significantly outperforms GPT-4o on the vast majority of reasoning-heavy tasks.
