OpenAI Launches ChatGPT Images 2.0

OpenAI has released ChatGPT Images 2.0, a major upgrade to its image generation capabilities that delivers greater precision, enhanced realism across diverse visual styles, improved multilingual text rendering, flexible aspect ratios, and the ability to act as an end-to-end visual thought partner.

openai Apr 21, 2026

OpenAI has unveiled ChatGPT Images 2.0, marking what the company describes as a new era of image generation. This major update brings a range of improvements to the image creation capabilities within ChatGPT.

Greater Precision and Control

ChatGPT Images 2.0 offers significantly improved accuracy and control over generated images, allowing users to achieve results that more closely match their creative intent.

Stylistic Sophistication and Realism

The updated model delivers enhanced fidelity across a wide variety of visual styles, including photography, illustration, manga, pixel art, and other artistic formats. It can produce cinematic portraits, candid street photography, high-fashion editorial imagery, surrealist compositions, and detailed comic pages with multi-scene continuity.

Stronger Multilingual Support

OpenAI's new image generation system demonstrates markedly improved performance across global languages and scripts. It can render text accurately in Japanese, Korean, Hindi, Chinese, Arabic, Thai, and many other languages, enabling the creation of multilingual posters, advertisements, manga, and branding materials with proper typography.

Flexible Aspect Ratios

The model supports a wide range of output formats, from horizontal banners and panoramic scenes to square and vertical compositions, accommodating diverse use cases from mobile screens to wide displays.

Enhanced Real-World Intelligence

ChatGPT Images 2.0 leverages up-to-date knowledge and contextual accuracy to transform information into clear, well-designed visual outputs. It can generate educational infographics, academic posters, design trend overviews, and data visualizations with structured reasoning and pedagogical clarity.

Visual Thought Partner

Acting as a visual thought partner, the thinking model can research, reason, transform source materials, and generate polished visuals end-to-end. This means users can turn rough inputs into cohesive creative assets-such as product mockups, branding concepts, café launch posters, bookmarks with print-ready production guides, and character reference sheets-with far less manual effort.

The feature is available to try directly in ChatGPT.