Amid rising competition between tech giants like developing advanced models capable of handling challenging tasks, OpenAI announced that it is testing new reasoning AI models, o3 and o3 small.

By January 2025, OpenAI CEO Sam Altman revealed that the company will roll out o3-mini, and then the full o3 model.

The objective of these cutting-edge models is to pool in new users and investments.

The o1 and o1-mini models, which OpenAI introduced in September, are intended to tackle challenging inquiries and reasoning tasks.

They set themselves apart from popular models like GPT-4o by resolving challenging issues like algebra and programming.

Compared to earlier models in science, coding, and math, the O1 models can reason through complex tasks and tackle more difficult challenges, the AI company claimed in a blog post.

OpenAI claims that its recently released o3 and o3 small models, which are presently undergoing internal safety testing, would be more potent than its earlier o1 versions.

Prior to the public release, which is scheduled to end on January 10, the GenAI pioneer announced that it is now accepting applications from outside researchers to test o3 models.

When OpenAI introduced ChatGPT in November 2022, it set off an AI arms race. In October, OpenAI closed a $6.6 billion investment round, aided by the company's increasing popularity and the introduction of new products.

Alphabet's competitors In an effort to take back the lead in the race for AI technology, Google earlier in December unveiled the second version of its AI model, Gemini.