Summary

  • OpenAI has launched o3 and o4-mini, AI systems that can use images in their reasoning process as well as search the web, run code and analyse files.
  • “They don’t just see an image, they think with it,” said OpenAI about the systems that it claims are its smartest models to date.
  • The models could revolutionise fields such as scientific research and education by zooming in on images to inspect details and rotating diagrams.
  • The models are also highly capable at coding and software engineering, in addition to other complex workloads and multi-step workflows.
  • The models are now available to ChatGPT Plus, Pro and Team users with Enterprise and Education customers gaining access next week.
  • This comes just two days after the unveiling of its GPT-4.1 model, indicating an acceleration of competition in the AI space.

By Michael Nuñez

Original Article