OpenAI has launched its next-gen image generation model ChatGPT Images 2.0. This model can generate better, more accurate and reference images based on prompts entered by users. Multilingual rendering and composition have been done in the new model. Reasoning capacity has also been included in it for more difficult tasks. ChatGPT Images 2.0 is currently rolling out on ChatGPT, Codex, and API. Let us know in detail about ChatGPT Images 2.0.
When will ChatGPT Images 2.0 be available?
OpenAI’s AccordingChatGPT Images 2.0 is currently available to all users of ChatGPT and Codex. Advanced capabilities, including reasoning-based thinking features, are available to ChatGPT Plus, Pro, and Business customers. Developers can access the model through the gpt-image-2 API. One thing to note is that the price will vary depending on the image quality and resolution. The new model supports output up to 2K resolution.
ChatGPT Images 2.0 Features
According to OpenAI, ChatGPT Images 2.0 offers more precise photo controls than previous-gen models, allowing users to create photos with detailed prompts. It can manage quite complex compositions with ease. This includes UI elements, dense text, and structured layout. Improvements have been made in terms of multilingual support. It displays non-English text more accurately, including languages like Hindi, Bengali, Chinese, Japanese, and Korean. Users can create posters, diagrams and infographics in languages.
According to the company, ChatGPT Images 2.0 provides better consistency across multiple styles, including photorealistic images, cinematic visuals, pixel art, and manga. Improvements have also been made in the areas of lighting, textures and composition. The latest image generation model also supports flexible aspect ratios from ultra-wide formats (3:1) to tall layouts (1:3).
With ChatGPT Images 2.0, OpenAI also develops thinking capabilities. According to the AI firm, this allows the model to perform tasks with more advanced reasoning capabilities than ever before. It can be combined with ChatGPT models with reasoning capabilities, allowing it to search the web for real-time information, verify results, and generate images from a single prompt.
