ChatGPT’s new Images 2.0 model is surprisingly good at generating text
TechCrunch
OpenAI's new Images 2.0 model demonstrates significant improvements in rendering legible text and complex visual elements within AI-generated images.Summary
OpenAI has introduced ChatGPT Images 2.0, a new image generation model that excels at rendering accurate text, a task that historically challenged diffusion-based models. By utilizing what OpenAI describes as "thinking capabilities," the model can better follow complex instructions, render diverse scripts including Japanese and Hindi, and generate multi-paneled layouts or marketing assets. While OpenAI has not confirmed the underlying architecture, the model’s ability to handle fine-grained details and small text at up to 2K resolution marks a notable advancement in generative AI quality.
(Source：TechCrunch)