Sam Altman's OpenAI Announces New o3 Models, Set To Launch In January 2025 (Screenshot:X/openai)
OpenAI announced the introduction of its latest reasoning AI models o3 and o3-mini on Saturday. These models have been designed to tackle complex problems and have already demonstrated exceptional performance in various benchmarks.
According to OpenAI CEO Sam Altman, the company plans to launch the o3 mini by the end of January, followed by the full o3 model. This move is expected to attract new investors and users as the more robust models are likely to outperform existing ones.
Today, we shared evals for an early version of the next model in our o-model reasoning series: OpenAI o3 pic.twitter.com/e4dQWdLbAD
— OpenAI (@OpenAI) December 20, 2024
The company said that the o3 model has shown remarkable results, outscoring OpenAI’s Chief Scientist in competitive programming and achieving impressive scores in coding tests and math contests.
Also, read| OpenAI Upgrades ChatGPT With Vision Capabilities And Festive “Santa Mode”
Specifically, it has excelled in SWE-Bench verified coding tests by 22.8% and scored 87.7% in the AIME 2024 math contest.
We also shared evals on Open AI o3-mini — a faster, distilled version of o3 which is optimized for coding, and the first version of o3 we expect to make available for use in early 2025. pic.twitter.com/0JfK7ER6HA
— OpenAI (@OpenAI) December 20, 2024
Also, OpenAI has skipped the o2 model to avoid potential copyright conflicts with a British company named O2. The company is now accepting applications from researchers to test the o3 models before their public release with the application process closing on January 10.
Also, read| Microsoft Quits OpenAI’s Observer Role Amid Growing Antitrust Issues
These new models are expected to be more powerful than their predecessors, o1 which were capable of reasoning through complex tasks and solving tough problems. The o3 and o3-mini models are currently undergoing internal safety testing.
