ChatGPT Images 2.5 is now rolling out as a new image generation model. It promises sharper details, more precise editing, and faster output for creative workflows.

The update comes at a time of massive scale. Every week, users create more than 3 billion images across ChatGPT Images and the GPT-Image models in the API. Against that backdrop, the company is positioning Images 2.5 as a meaningful step forward.

What’s New in Images 2.5

According to the announcement, Images 2.5 produces more natural lighting and richer textures. Additionally, it is better at preserving the subjects in reference photos. As a result, edits carry through more consistently across multiple turns.

Speed has also improved. Image generation latency has been reduced by up to 50% compared with Images 2.0. Consequently, users can generate images and refine concepts more quickly than before.

New Product Features in ChatGPT

Alongside the model update, several product features are being introduced to give users more control over the creative process:

• Sketch: A new feature that lets users draw directly in ChatGPT as a reference for a final image.
• Templates: These make it easier to start creating across popular image formats, such as flyers and product photos.
• Comments on images: Users can now place comments directly on images for more focused editing.
• Shareable prompts: Users can share prompts they’ve used, letting others try the same ideas with their own photos and details.

Furthermore, Images 2.5 is available to all ChatGPT, ChatGPT Work, and Codex users. It works across desktop, mobile, and web platforms.

API Access for Developers

For developers, two new models are being introduced in the API:

1. GPT-Image-2.5 Flare: Brings the same improvements in quality, editing, and speed.
2. GPT-Image-2.5 Sunburst: Offers an extra level of precision for detailed creative work, though with longer generation times.

Therefore, teams can choose between speed and precision depending on their project needs.

Image Fidelity and Reference Photos

The most meaningful images are grounded in real-life people, places, and memories. In line with this, Images 2.5 is better at working from reference photos to transform familiar subjects across new settings, visual styles, and compositions.

As a result, image subjects look more recognizable. Lighting and textures feel more natural, and distinctive features are more likely to carry through the final output.

Moreover, this same fidelity extends to API-based workflows. For teams building reference-led applications, variations can now stay more reliably anchored to the original source image.

Looking Ahead

Overall, the rollout of Images 2.5 signals a broader push toward more controllable, higher-fidelity AI image generation, both for everyday ChatGPT users and for developers building on the API.

Related post

View all