The revamped tool, featuring enhanced ‘thinking capabilities,’ is named ChatGPT Images 2.0 and aims to produce more refined images while improving the system’s comprehension of user commands.
The new iteration of the image generator utilizes OpenAI’s latest GPT Image 2 model and brings in advanced thinking features. This allows the system to delve deeper into prompts, reason through their structure, and leverage web information for image creation. Additionally, it can generate visual explanations using uploaded documents.
These advanced thinking features are now accessible to subscribers of ChatGPT Plus, Pro, Business, and Enterprise plans.
On X, the AI firm stated, “Introducing
ChatGPT Images 2.0, a cutting-edge image model capable of tackling complex visual tasks and delivering precise, immediately applicable visuals, with improved editing, richer layouts, and thinking-level intelligence.”
Introducing ChatGPT Images 2.0
A cutting-edge image model that can tackle complex visual tasks and deliver precise, immediately applicable visuals, with improved editing, richer layouts, and thinking-level intelligence.
Video made with ChatGPT Images pic.twitter.com/3aWfXakrcR
— OpenAI (@OpenAI) April 21, 2026
A key feature is the capability to produce up to eight images simultaneously from one prompt when thinking is activated. OpenAI mentioned that the tool can keep consistent elements like characters, objects, and styles throughout all generated images. This makes it particularly advantageous for producing content such as manga pages, design layouts, or social media graphics.
In a blog post, OpenAI remarked, “When a thinking or pro model is selected in ChatGPT, Images 2.0 can search the web for real-time information, generate multiple distinct images from a single prompt, and verify its outputs. With thinking, the model can take on even more of the heavy lifting between concept and image, especially when accuracy, current information, consistency, and visual cohesion are of utmost importance.”
Alongside new features, ChatGPT Images 2.0 enhances image quality and detail. The tool can now create visuals with resolutions up to 2K and supports a broader variety of aspect ratios. It also excels at capturing specific visual styles like pixel art, cinematic scenes, and manga illustrations.
Furthermore, OpenAI has indicated that the model shows significant progress in generating text within images across multiple languages, including Hindi, Bengali, Japanese, Korean, and Chinese, enhancing accessibility for global users.
The upgraded image generator is currently being rolled out to all ChatGPT and Codex users. OpenAI introduced its image tool last year and has been continuously refining it in light of increasing competition from other tech companies creating similar AI-powered creative solutions.
