OpenAI has introduced ChatGPT Images 2.5, its latest image-generation model designed to give users more control over the creative process. The new model focuses on four major areas. These include image quality, generation speed, editing accuracy and consistency across multiple edits.
The update comes as AI image generation becomes a bigger part of ChatGPT. OpenAI says users now create more than 3 billion images every week across ChatGPT Images and the GPT-Image models available through its API.
ChatGPT Images 2.5 Is Up To 50% Faster
ChatGPT Images 2.5—faster, sharper, smarter, with better tools for creating whatever you can dream of.
– Faster image generation to keep your ideas flowing
– Improved fidelity for more natural, recognizable images
– Consistent details across multiple edits
– Comment-based… pic.twitter.com/SjHITKJUSV
— OpenAI (@OpenAI) September 8, 2026One of the biggest improvements is generation speed. OpenAI says Images 2.5 reduces image-generation latency by up to 50 percent compared with Images 2.0. This should make the process of generating and refining images considerably quicker. Users can create an image, make changes and continue refining the result without waiting as long between generations.
The model also produces more natural lighting and richer textures. OpenAI says it has improved the way Images 2.5 handles reference photos, allowing subjects to remain more recognisable when moved into different environments, styles or compositions.
More Precise Image Editing
Editing is another major focus of the update. Images 2.5 is designed to change only the part of an image that the user asks to modify while keeping the remaining details intact. For example, users can change a background, product or individual element without having the entire composition change unexpectedly. This is particularly useful for product images, marketing creatives and other visual work where consistency matters.
The model is also better at following instructions through several rounds of editing. Earlier changes are more likely to remain intact as users continue modifying the same image.
Better Results From Complex Prompts
Images 2.5 also brings improvements in understanding complex visual instructions. OpenAI says the model can handle more complicated layouts and real-world information with greater accuracy. It also supports more complex compositions, including transparent backgrounds. Style control has been improved as well, allowing generated images to follow a requested visual direction more closely.
This could be useful for creators and businesses working on a series of images that need to maintain the same visual identity, composition and branding.
Sketch Lets Users Draw Their Ideas
Some ideas are easier to draw than describe.
Use Sketch to draw right in ChatGPT and show it exactly what you have in mind.
Just type “@ Sketch” in ChatGPT. pic.twitter.com/gx6fw0xy3K
— OpenAI (@OpenAI) September 8, 2026OpenAI has also introduced a new feature called Sketch. Instead of describing everything through text, users can draw directly inside ChatGPT and use that sketch as a visual reference. A rough drawing can be used to describe the layout of a room, an outfit, a product or almost any other visual idea. ChatGPT can then turn that rough concept into a finished image. Users can access the feature by typing @Sketch in ChatGPT.
Templates Make Image Creation Easier
Have an idea but need help getting started?
Use templates for popular image formats like posters or merch, then add your message, design elements, or style to make it your own. pic.twitter.com/SD8cOyZT7a
— OpenAI (@OpenAI) September 8, 2026Images 2.5 also introduces templates for common creative formats. Users can start with formats such as posters and merchandise instead of beginning with a completely blank prompt. They can then add the information, design elements and visual style they want. This could make the image-generation process easier for users who have an idea but are unsure how to structure the prompt.
OpenAI has also added the ability to place comments directly on generated images. These comments can be used to tell ChatGPT exactly which part needs to be changed. Users can also share the prompt used to create an image, allowing others to recreate the concept with their own photos and details.
GPT-Image-2.5 Flare And Sunburst For Developers
Images 2.5 is here.
I don't think it can solve super difficult math problems, but it is really good and we hope you enjoy it.https://t.co/LdZhGF5AcZ
— Sam Altman (@sama) September 8, 2026OpenAI is also bringing the Images 2.5 improvements to developers through two new API models.
GPT-Image-2.5 Flare is the default choice for most applications. OpenAI says it provides higher-quality images than GPT-Image-2 while offering 50 percent lower latency.
GPT-Image-2.5 Sunburst is aimed at premium visual workflows where more precise control is important. It uses longer generation times to provide additional precision for detailed creative work.
ChatGPT Images 2.5: Key Specifications
| Feature | Details |
|---|---|
| Model | ChatGPT Images 2.5 |
| Generation Speed | Up to 50% lower latency vs Images 2.0 |
| Image Quality | Sharper details, richer textures, natural lighting |
| Reference Images | Improved subject and feature preservation |
| Editing | More precise, targeted edits |
| Multi-Turn Editing | Better consistency across multiple edits |
| Complex Layouts | Improved support, including transparent backgrounds |
| Sketch | Available through @Sketch |
| Templates | Posters, merchandise and other formats |
| Image Comments | Supported |
| Prompt Sharing | Supported |
| Availability | ChatGPT, ChatGPT Work and Codex |
| Platforms | Desktop, mobile and web |
| API Models | GPT-Image-2.5 Flare and Sunburst |
Availability
ChatGPT Images 2.5 is rolling out today to all ChatGPT, ChatGPT Work, and Codex users across desktop, mobile, and web.
We’re also introducing two new models in the API: GPT-Image-2.5 Flare brings the same quality, editing, and speed improvements, while GPT-Image-2.5 Sunburst…
— OpenAI (@OpenAI) September 8, 2026OpenAI says ChatGPT Images 2.5 is rolling out to ChatGPT, ChatGPT Work and Codex users across all tiers on desktop, mobile and web. The two new API models, GPT-Image-2.5 Flare and GPT-Image-2.5 Sunburst, are also available for developers. OpenAI is also retaining its existing image safety measures. The company says Images 2.5 uses prompt and image checks, along with C2PA metadata and invisible watermarking to help identify images created with its tools.
Conclusion
ChatGPT Images 2.5 is more than a simple image-quality upgrade. OpenAI is also making the entire creative workflow easier. Faster generation should make experimentation quicker, while better reference fidelity and multi-turn editing can help users maintain consistency.
Features such as Sketch, templates and image comments add another layer of control. For developers, the new Flare and Sunburst models extend those improvements into API-based creative workflows. Overall, Images 2.5 looks like a significant step toward making AI image generation more practical for both everyday users and professional creators.
Follow Us: Facebook | X | Instagram | YouTube | Pinterest
