OpenAI Launches ChatGPT Images 2.0 with Enhanced Capabilities
Original: OpenAI Beefs Up ChatGPT’s Image Generation Model
Why This Matters
Represents major advancement in AI image generation quality and multilingual capabilities
OpenAI released ChatGPT Images 2.0 on Tuesday, featuring improved image generation that can create multiple images from single prompts, render text more accurately, and generate content in non-English languages like Chinese and Hindi.
The new model leverages ChatGPT's reasoning capabilities to search the internet for recent information and generate multiple images simultaneously. It has an updated knowledge cutoff date of December 2025 and offers more customizable aspect ratios from 3:1 wide to 1:3 tall. Testing showed significant improvements in text rendering accuracy compared to previous versions that often produced malformed characters. The model can generate complex outputs like infographics with accurate weather data and detailed location drawings. When tested with non-English content, it successfully created a detailed Timothée Chalamet-themed collage with Chinese text and cultural elements. The release is available globally for ChatGPT users, with enhanced features for paying subscribers.