GPT Image 2.5 in AI Image Generation Action In Workflows

GPT Image 2.5 is now live in the AI Image Generation action. Two variants ship together, Flare and Sunburst. Open the model dropdown, pick the one that fits the job, and generate.

The default pick for most use cases. Flare produces higher quality output than GPT Image 2 at up to 50% lower latency, which makes it the right fit for creator and social content, product visuals, visual search, quick prototyping, and high volume generation.

Built for premium visual work that needs tighter control across edits. Use it for campaign creative and polished product imagery.

Reference photos carry much further than before. Subjects stay recognizable across new settings, styles, and compositions, lighting and textures look natural, and distinctive features carry through instead of getting lost. Variations stay anchored to the original source, so a series holds together rather than drifting image to image.

You can also swap a single element, such as a product, a background, or a piece of copy, while the subject, composition, and brand treatment stay intact. That turns image generation into something you can re-version and localize inside a workflow instead of rebuilding each asset from scratch.