OpenAI launched ChatGPT Images 2.0 on Tuesday, a new image generation AI model with significantly improved capabilities, according to both Wired and TechCrunch.
According to Wired, the model can generate more than one image from a single prompt, such as “an entire study booklet,” and can output text in non-English languages including Chinese and Hindi. The release is available globally for ChatGPT and Codex users, with a more powerful version offered to paying subscribers.
TechCrunch highlighted the model’s notable improvement in text generation within images. The publication noted that just two years ago, AI image models struggled to create readable text, citing an example where a Mexican restaurant menu would contain misspellings like “enchuita,” “churiros,” “burrto,” and “margartas.” According to TechCrunch, when testing the new ChatGPT Images 2.0 with a Mexican food menu prompt, the model “creates something that could immediately be used in a restaurant without customers noticing that something’s off.”
Wired also noted that new image model releases can drive user engagement, particularly when social media users adopt trending uses. The publication referenced Google’s Nano Banana model launch last year, which gained traction when users posted hyperrealistic figurines, and earlier ChatGPT Images releases that sparked AI-generated caricature trends.