OpenAI has announced the launch of the updated ChatGPT Images, which is now four times faster at generating images. This release follows the introduction of GPT-5.2 and aims to compete directly with Google’s Nano Banana Pro, similar to how GPT-5.2 responds to Gemini 3 Pro.
According to OpenAI’s press release, ChatGPT Images is designed to better adhere to user intent, allowing for precise edits down to small details. It modifies only the requested elements while preserving the overall integrity of the image across various inputs and outputs.
The new features are expected to provide results that align closely with user expectations, positioning ChatGPT as a versatile creative tool capable of practical edits and imaginative reworkings. The model has improved capabilities for adding, subtracting, combining, blending, and transposing elements, ensuring that desired changes are made without compromising the unique aspects of the image.
Users will find preset styles and ideas within the ChatGPT Images feature, eliminating the need for written prompts. The model’s ability to follow instructions has also been enhanced, allowing for more accurate edits and complex original compositions while maintaining the intended relationships between elements. Improvements in text rendering and the depiction of small faces are also part of this update.
The new ChatGPT Images is currently being rolled out to users worldwide.
—
