Tuesday, 22 April 2025
29 C
Singapore
33.4 C
Thailand
21.2 C
Indonesia
29.4 C
Philippines

OpenAI introduces powerful new AI models with advanced image reasoning

OpenAI’s new o3 and o4-mini AI models bring powerful image reasoning and full tool access to ChatGPT Plus, Pro, and Team users.

OpenAI has just unveiled two new AI models to revolutionize the way you work with artificial intelligence. Known as o3 and o4-mini, these models promise better performance and more flexible tools. What makes them especially interesting is their ability to understand and use images in their thinking process.

Powerful reasoning with images

The o3 model is called OpenAI’s most powerful reasoning model so far. It’s designed to be smarter and more capable than previous versions. Meanwhile, the o4 mini is a smaller, faster option that is still impressively effective, mainly because of its size and cost.

What’s special about both models is their ability to reason using images. You can show them sketches, diagrams, or even whiteboard notes, and they can include those visuals in their thought process. They won’t just look at the images—they’ll interact with them by zooming in, rotating, or focusing on specific areas to better understand your questions.

This kind of image reasoning could be helpful in many fields. For example, if you’re a teacher, designer, or engineer, you might want to show the AI a drawing or a plan and ask questions about it. These models can now offer deeper insights because they actually “think” with those images in mind.

More tools available to more users

The new reasoning models will also be able to use all the tools of ChatGPT, including web browsing, image generation, and more. These tools help the AI look up current information or create pictures based on your ideas.

If you’re using ChatGPT with a Plus, Pro, or Team plan, you can try these tools with the o3, o4-mini, and o4-mini-high models starting today. Users of the o3-pro model will get access in a few weeks.

At the same time, older versions like o1, o3-mini, and o3-mini-high will be removed from these paid plans. This change is part of OpenAI’s goal to focus on its newest and most advanced models.

A week of exciting updates from OpenAI

These announcements come just after OpenAI introduced another major update earlier this week. On Monday, the company revealed GPT-4.1, its latest flagship model. This version follows GPT-4o and is expected to be even more capable.

With all these changes, it’s clear that OpenAI is pushing hard to make its tools more helpful, especially regarding reasoning and understanding the world more like a person would.

So, if you’re using AI for work, study, or creativity, these new tools could give you better results and more freedom to explore new ideas—with words and now with images, too.

Hot this week

ChatGPT trend raises privacy concerns with photo-based location searches

People use ChatGPT to identify photo locations, raising privacy concerns as new AI tools make “reverse location search” easier than ever.

OpenAI introduces Flex processing to cut AI costs for slower tasks

OpenAI launches Flex processing, cutting AI usage costs by 50% for non-urgent tasks using o3 and o4-mini models with slower response times.

Apple chooses Samsung for foldable iPhone screen due to crease-free display

Apple is choosing Samsung Display for its foldable iPhone to avoid screen creases. A major iPhone redesign is set for 2026.

Chinese tech companies race to expand AI services using new open-standard protocol

Chinese tech firms race to adopt MCP, boosting AI agent use and shaping the future of smart services in payment, maps, and cloud tools.

ASUS introduces the first smart band with fingertip blood pressure and ECG tracking

ASUS launches VivoWatch 6 Aero, the first smart band with fingertip blood pressure and ECG tracking, powered by advanced health AI.

LG reveals a smart monitor on wheels—but it still needs to be plugged in

LG’s new 31.5" Smart Monitor Swing rolls on wheels and runs webOS, but it still needs a power plug and costs around US$740.

Amazon sets new date for first Project Kuiper satellite launch

Amazon’s first Project Kuiper launch is now set for April 28. The launch will use an Atlas V rocket to carry 27 satellites and bring global broadband.

Razer quietly resumes laptop sales after a sudden pause in the US

Razer resumes some US laptop sales after a sudden halt, with limited models available and no explanation from the company.

AMD optimizes models for faster generative AI performance with Amuse 3.0

AMD announces optimised models for faster AI performance with Amuse 3.0, featuring enhanced photo filters, video diffusion, and more.

Related Articles

Popular Categories