OpenAI Launches ChatGPT Images 2.0 With Enhanced Multilingual Text and Visual Rendering
OpenAI has unveiled ChatGPT Images 2.0, a significantly upgraded image generation model that demonstrates remarkable improvements in rendering text and creating complex visual content. The company announced the new model on April 21, 2026, through a livestream event that detailed capabilities well beyond what previous versions could achieve.
The most striking advancement is the model's ability to generate multilingual text within images. Where earlier versions struggled to render accurate text of any kind, Images 2.0 can now create full infographics, slides, maps, and even manga-style artwork with text integrated seamlessly. According to reports, the model performs these tasks "seemingly flawlessly," marking a substantial leap forward in addressing one of the most persistent limitations in AI image generation. The model also demonstrates improved instruction-following overall, with better handling of colors, lighting, and detailed compositional elements.
This release comes just months after OpenAI introduced GPT-Image-1.5 in December 2025, which itself brought notable improvements to instruction following and visual quality. The rapid succession of upgrades reflects the accelerating pace of development in generative AI capabilities. The new model also introduces "thinking capabilities" that allow it to search the web and generate multiple sophisticated variations from a single user prompt, expanding its practical applications.
The rollout includes new API access and enterprise pricing options, making the technology available to developers and businesses looking to integrate advanced image generation into their products and services. This expansion signals OpenAI's push to make Images 2.0 a cornerstone tool across both consumer and enterprise applications.
However, testing has revealed some remaining limitations. While the model excels at English text rendering, it still encounters challenges with non-English languages, suggesting that multilingual text generation, though improved, remains an area for future refinement. These constraints notwithstanding, Images 2.0 represents a meaningful advancement in closing the gap between human-created and AI-generated visual content, particularly for professional and creative use cases that require integrated text and complex visual design.