OpenAI has released ChatGPT Images 2.0, a new image generation model that shows dramatically improved text rendering compared to previous generations. The model now generates legible text suitable for immediate commercial use, whereas prior models like DALL-E 3 produced nonsensical spellings that made images impractical. OpenAI suggests the improvement comes from using autoregressive models instead of traditional diffusion approaches, though declined to confirm technical details.
Models
ChatGPT’s new Images 2.0 model is surprisingly good at generating text
OpenAI's Images 2.0 shifts from diffusion to autoregressive models, finally generating legible text in images—solving a critical blocker for commercial image generation.
Tuesday, April 21, 2026 12:00 PM UTC2 MIN READSOURCE: TechCrunchBY sys://pipeline
Tags
models
/// RELATED
ProductsApr 21
OpenAI’s updated image generator can now pull information from the web
OpenAI adds web-aware reasoning to ChatGPT Images 2.0, letting it generate up to eight coherent images at 2K resolution informed by web search—a reasoning-powered upgrade exclusive to paid subscribers.
ProductsApr 21
OpenAI's ChatGPT Images 2.0 is here and it does multilingual text, full infographics, slides, maps, even manga — seemingly flawlessly
OpenAI's ChatGPT Images 2.0 leaps beyond simple visuals to generate complex professional content—multilingual infographics, slides, maps, and manga—with native text rendering and layout control.