【Watch Project Power Online】

OpenAI's advanced reasoning model has arrived.

Initially dubbed "Project Strawberry," the model was released to the public as "o1" on Thursday. As expected, o1 is able to "think" (process) before answering a prompt, which differs from previous large language models that use predictive text to form their responses. "We've developed a new series of AI models designed to spend more time thinking before they respond," said OpenAI in the announcement.

This Tweet is currently unavailable. It might be loading or has been removed.

This Tweet is currently unavailable. It might be loading or has been removed.

In a demo shared on the OpenAI site, the video shows a user asking o1 in ChatGPT to "write a grammatictlly [sic] correct sentence without using any letters more than once." The model then uses its new reasoning skills to weigh options and evaluate the sentence. According to the demo, after 39 seconds, o1's cheeky response to the query is "go fix my bed."

Another video features Research Lead Jerry Tworek asking o1 to solve a logic puzzle about the ages of a prince and princess. The model is shown decoding the problem, by understanding the different variables and determining which equations are needed to solve the problem, and goes step by step answering the age equation.

As demonstrated from the demos, o1 works best for solving complex math problems in the realm of science, physics, and coding. "o1 can be used by healthcare researchers to annotate cell sequencing data, by physicists to generate complicated mathematical formulas needed for quantum optics, and by developers in all fields to build and execute multi-step workflows," said the announcement.

That also means the o1 preview comes with certain limitations. It can't browse the web or take in uploaded files and images. For tasks like those, OpenAI says, GPT-4o is still the best model to use.

• Elon Musk is suing OpenAI and Sam Altman again after dropping his previous lawsuit
• OpenAI CEO Sam Altman was fired for 'outright lying,' says former board member
• OpenAI launches new internal safety team with Sam Altman in control

In a section about safety, OpenAI said the U.S. and UK Safety Institutes were granted early access to testing o1, per the previous mandates from President Biden's AI executive order.

OpenAI's o1 model is available through the developer AI and ChatGPT Plus users in preview. When you go to the ChatGPT app, you can select o1 in the dropdown menu.