OpenAI has introduced ChatGPT Images 2.5, an advanced image-generation model aimed at enhancing detail, editing precision, and reducing generation time. The new model is capable of producing images with improved natural lighting and richer textures, while maintaining consistency with subjects from reference photos. In addition to the model, OpenAI has rolled out new tools within ChatGPT, including Sketch, templates, image comments, and prompt sharing. ChatGPT Images 2.5 is now accessible to users of ChatGPT, ChatGPT Work, and Codex on desktop, mobile, and web platforms.
OpenAI Unveils Images 2.5 with Enhanced Performance
OpenAI has introduced Images 2.5, a new image generation model that boasts up to 50% lower latency compared to its predecessor, Images 2.0. This improvement allows users to create and update images more quickly.
The model features enhanced capabilities for interpreting detailed image instructions. According to OpenAI, it can better preserve intricate details from reference photos while allowing changes in setting, style, or composition. Additionally, Images 2.5 is designed to manage more complex layouts, support transparent backgrounds, and accommodate diverse visual styles.
New Sketch feature lets users draw image ideas
ChatGPT Images 2.5 has also introduced templates for some common image formats. OpenAI has also introduced Sketch, a new feature that allows users to draw directly in ChatGPT and use the drawing as a reference for image generation.
The company says users can access the feature by typing @Sketch in ChatGPT. Users can make a rough drawing of a room, outfit or another idea and then describe the desired style and other details. Users can start with formats such as posters and merchandise designs instead of beginning with a blank prompt. They can then add details about the information, design elements and visual style they want. Another new feature allows users to share the prompt used to create an image. When an image is shared, users can choose to include its prompt. Other people can then use the same idea with their own photos and details. OpenAI has highlighted an example involving an image that turns a person’s photo into an 1980s-style portrait.
ChatGPT Images 2.5 Now Available to All Users
OpenAI has announced the rollout of ChatGPT Images 2.5, which is now accessible to ChatGPT, ChatGPT Work, and Codex users across all subscription tiers. This update is available on desktop, mobile, and web platforms. Additionally, the company has released two new models, GPT-Image-2.5 Flare and GPT-Image-2.5 Sunburst, through its API, expanding the tools available for developers and users alike.