Newsbite analysis · AI, Programming, Tools · New Product, Software Release
PCMag
ChatGPT Adds Images 2.5 Model, New Feature Turns Doodles Into AI Photos
OpenAI has released the Images 2.5 model for ChatGPT, significantly reducing image generation latency by 50% and enhancing output quality, particularly in natural lighting and textures. The update incorporates a new Sketch tool, allowing users to create high-quality images from their doodles by simply sketching and providing a brief description. Additionally, the model features Templates for quicker creative processes, enabling swift generation of specific formats like posters. This upgrade arrives shortly after the announcement of GPT-6 Astra, further expanding ChatGPT's capabilities.
Act nowConfidence: HighImmediate horizon
Published 9 Sept 2026, 14:55 UTC2 min read
Lead image — PCMag
Brief
NewsBite reading:OpenAI Updates ChatGPT with Images 2.5 and Sketch Feature for Doodles
TL;DR
- Images 2.5 model reduces image generation latency by 50%
- New Sketch tool turns user doodles into detailed AI images
- Templates for faster creation are now available
- Improvements in natural lighting and texture fidelity
- Updated model follows the release of GPT-6 Astra
Interesting facts
- The Sketch tool utilizes user sketches paired with prompts for enhanced output.
- New capabilities aim to streamline creative processes for users.
What changed
OpenAI launched the updated Images 2.5 model and the new Sketch tool for converting doodles into high-quality images.