OpenAI has officially announced a new image generation model, the ChatGPT Images 2.0.
OpenAI claims that the updated model now offers better instruction following, allowing it to place objects more accurately and render detailed elements such as small text, icons, and user interface components.
It also supports multiple aspect ratios, making it suitable for different formats like social media posts, banners, and presentations.
Images 2.0 also improves multilingual support, especially for non-Latin languages, including Japanese, Korean, Chinese, Hindi, and Bengali. Text in generated images is more consistent and readable, even in complex layouts.
The model also shows better handling of different visual styles, it produces more consistent lighting, textures, and composition, aiming for outputs that look less artificial.
A new “thinking” capability is available when using higher-tier models. This allows the system to process more complex tasks, search for up-to-date information, generate multiple images from one prompt, and check outputs for accuracy.
ChatGPT Images 2.0 is available starting today in ChatGPT, Codex, and through the API using the gpt-image-2 model. Advanced features are offered in paid plans.
OpenAI notes that the model still has limitations, including challenges with highly detailed scenes, complex diagrams, and objects that require precise physical accuracy.
ManilaShaker is a tech media producing insightful and helpful content for our local and growing international audience. Our goal is to create a premier Philippine digital consumer electronics resource that provides the most objective reviews and comparisons globally.
