These two new models are already available to ChatGPT Plus, Pro, and Team subscribers. — AFP
OpenAI has just unveiled two new artificial intelligence models, o3 and o4-mini, which are already available to paying subscribers of ChatGPT. The company is billing these new releases as reasoning models showcasing their latest developments, with capabilities including web search, memory referencing, improved coding, and image analysis.
The company behind ChatGPT has announced the arrival of two new AI models marking a significant advance in the reasoning and autonomous capabilities of the tools available to users via their platform. What OpenAI has named o3 is emerging as the company's most advanced reasoning model to date. According to its announcement, the model excels in complex fields such as mathematics, programming, science, and visual analysis.
While this model is of course able to analyze photos, it can also analyze diagrams, blurry and low-quality images, and even tables and hand-drawn sketches. Even more impressive, it can "decide" to adjust these elements by rotating or zooming in on the image to extract relevant information. An even more powerful version, called o3-pro, is already in the works and should soon be available to Pro subscribers.
Meanwhile, the o4-mini model appears to be a more compact and faster version, again particularly effective in tasks related to mathematics, coding, and visual analysis. Although lighter, the o4-mini outperforms its predecessor, the o3-mini, in many areas.
These two new models are already available to ChatGPT Plus, Pro, and Team subscribers. While Pro subscribers enjoy virtually unlimited access, ChatGPT Plus and Team subscribers are subject to usage quotas. These are in addition to other models that are still operational, including GPT 4o, which can process text, images, and audio, and GPT 4.5, which will soon be replaced by GPT 4.1.
Don't forget, regardless of the model used, it is important to know how to protect your personal data on ChatGPT. – AFP Relaxnews
