OpenAI has released ChatGPT Images 2.0, an updated image generation model that improves text rendering, adds multilingual support, and enhances visual reasoning. The new version builds on the original model with higher accuracy in text integration and broader language coverage. It targets professionals needing precise visuals for design, education, and content creation.
The upgrade addresses a key limitation in earlier versions: inconsistent text rendering in generated images. OpenAI reports that Images 2.0 reduces errors in text placement and clarity, making it more reliable for tasks requiring embedded text. It also expands language support beyond English, including full compatibility with Norwegian and Ukrainian.
Visual reasoning capabilities have been refined to better interpret and generate complex scenes. The model now handles detailed prompts with improved context awareness, reducing mismatches between input and output. OpenAI states this is part of a broader push to make AI-generated images more practical for real-world use.
The release follows months of testing with select users and developers. OpenAI has not disclosed specific performance metrics but claims significant improvements in speed and output quality. The update is available to ChatGPT Plus and Enterprise users starting today.
Industry analysts note that Images 2.0 reflects growing competition in AI image generation, with rivals also refining text integration and multilingual features. OpenAI’s move underscores the technology’s expanding role in creative and professional workflows.
Source: openai.com