OpenAI ships ChatGPT Images 2.5
OpenAI released ChatGPT Images 2.5 to all ChatGPT tiers on Monday, claiming up to 50% lower generation latency than Images 2.0 and adding two new image models to its API for developers.
OpenAI has released ChatGPT Images 2.5, an updated image generation and editing model, alongside two new image models in its API.
The company says users create more than 3 billion images a week across ChatGPT Images and the GPT-Image API models. The new model is available to all ChatGPT, ChatGPT Work, and Codex users on desktop, mobile, and web.
OpenAI describes Images 2.5 as producing more natural lighting and richer textures, better preserving subjects from reference photos, and following editing instructions more reliably across multiple turns without quality degradation. It also claims better handling of complex layouts, including transparent backgrounds, and image generation latency reduced by up to 50% compared with Images 2.0.
Three ChatGPT features ship with it: Sketch, invoked by typing "@Sketch," which lets users draw in ChatGPT as a visual guide for a generated image; templates for common formats such as "Poster" and "Merch"; and comments placed directly on images for targeted edits. Users can now also share the prompt behind an image so others can rerun it with their own inputs.
For developers, GPT-Image-2.5 Flare is positioned as the default choice, delivering what OpenAI says are higher-quality images than GPT-Image-2 at 50% lower latency. GPT-Image-2.5 Sunburst targets more detailed creative work at the cost of longer generation times.
OpenAI's announcement does not include benchmark results, API pricing, or a rollout timeline for the API models.