Following the recent launch of GPT-5.2, OpenAI today introduces an enhanced version of ChatGPT Images, now generating images up to four times faster. This upgrade positions ChatGPT Images as a strong competitor to Google's Nano Banana Pro, just as GPT-5.2 directly challenges Gemini 3 Pro.
According to OpenAI’s press release, ChatGPT Images now "adheres to your intent more reliably," especially when making edits. The system modifies only the requested elements while preserving all other aspects across initial inputs, outputs, and subsequent adjustments.
This enhancement aims to deliver "results that match your intent," transforming ChatGPT Images into a versatile creative studio in your pocket. It excels in practical edits and imaginative transformations by effectively adding, removing, combining, blending, and transposing image elements without compromising what makes the image unique.
The updated ChatGPT Images feature includes preset styles and ideas that users can select without writing prompts. The model also follows instructions more accurately, enabling more detailed edits and intricate original compositions where the relationships between elements are maintained as intended. Additionally, improvements have been made in text rendering and the depiction of small faces.
The new ChatGPT Images capability is currently rolling out globally to all users.
Source
