ChatGPT now lets you create and edit images faster and more reliably

After upgrading ChatGPT with its latest GPT-5.2 model last week, OpenAI has rolled out a major improvement for the chatbot’s image generation capabilities, positioning it as a strong competitor to Google’s Nano Banana Pro. Powered by OpenAI’s new flagship image generation model, the latest version of ChatGPT Images promises up to four times faster image generation and far more accurate, reliable results that closely follow user instructions.

OpenAI says the new image generation model performs better, both when users generate images from scratch and when they edit existing photos. It preserves important details across edits while giving users precise control over changes. Users can add, subtract, combine, blend, and transpose elements while editing, and even add stylistic filters or perform conceptual transformations.

The company claims that the model “follows instructions more reliably than” the initial version, enabling “more precise edits as well as more intricate original compositions, where relationships between elements are preserved as intended.” Text rendering has also improved substantially, and the model can now handle denser, smaller text.

A dedicated workspace for ChatGPT Images

To further improve user experience, OpenAI has also given image generation a dedicated space within ChatGPT. In addition to generating images directly in the chat interface, users can now access ChatGPT Images from the sidebar in the mobile app and on the web. This new workspace includes preset filters and prompts to help users get started, and OpenAI says it will regularly update these presets “to reflect emerging trends.”

The new ChatGPT Images model is rolling out to all ChatGPT users across platforms. OpenAI notes that while the “model shows clear improvements across a range of cases,” it has some limitations in some languages. Its ability to generate a few art styles has also “regressed from the previous version,” and the model has “difficulty maintaining the exact identity of every person” when there are many people in an image.

Despite these limitations, the new ChatGPT Images model marks a major leap forward for AI-powered image creation. With faster generation, more precise edits, and a dedicated workspace to streamline creation, users now have a more versatile tool for creating and editing images. OpenAI says that this update “is a meaningful step forward with more to come, from finer-grained edits to richer, more detailed outputs across languages.”