OpenAI has rolled out an updated image generation model for ChatGPT, making it four times faster and better at following user instructions. The upgrade includes improved editing capabilities and enhanced text rendering. This comes shortly after the release of GPT-5.2 and amid competition from Google's Gemini.
Following the release of GPT-5.2 last week, OpenAI began rolling out a new image generation model on December 16, 2025. The updated ChatGPT Images is four times faster than its predecessor, addressing delays often experienced during peak times, especially for non-ChatGPT Plus users.
The model excels at following instructions, particularly for edits. Users can now ask it to add, subtract, combine, blend, or transpose elements in generated images. It also offers better text rendering, capable of handling denser and smaller text more effectively—a challenge for many image models.
As part of the update, OpenAI added a dedicated Images section to the ChatGPT sidebar, featuring preset filters and prompts for inspiration. This rollout arrives amid a surge in Google's Gemini usage, boosted by Nano Banana Pro, which grew from 450 million users in July to 650 million in October. Google recently limited free users to two image generations per day. OpenAI's ChatGPT, bolstered by this image tool, reaches 800 million users.
"We believe we’re still at the beginning of what image generation can enable," OpenAI said. "Today’s update is a meaningful step forward with more to come, from finer-grained edits to richer, more detailed outputs across languages."
The upgrade is available to all users, with the option to access the older GPT-4o version through a custom GPT.
