What happened
OpenAI released ChatGPT Images 2.5, with new API models gpt-image-2.5-sunburst and gpt-image-2.5-flare. The models handle multiple instruction turns, respond faster, and better preserve reference photo subjects.
Why it matters
Previous versions lacked such refined multi-turn editing and reference preservation. Now users can pass reference images and edit with more precision, as demonstrated with a raccoon scientist added to a chart.
What to watch
Whether Sunburst's precision advantage holds in practice. The distinction between Sunburst and Flare suggests workflows may choose based on editing needs versus speed.
WHO IT HITSDevelopers and creative professionals who use OpenAI's image generation APIs will be affected, as they can now choose between precision (Sunburst) and speed (Flare) for their workflows.
Ask the AI about this article →
Summaries like this, in your inbox every morning.
OpenAI's release of ChatGPT Images 2.5 comes as image generation becomes a core feature of AI platforms. The mention that over 3 billion images have been created underscores the scale of adoption. The new models address common pain points: following complex instructions over multiple turns and preserving the identity of subjects in reference photos. The choice between Sunburst and Flare offers a trade-off between editing precision and speed, catering to different user needs. This release may intensify competition among AI image generators, but the outcome hinges on whether these improvements meet user expectations in real-world tasks. As developers like Simon Willison integrate these models into tools, the practical implications for creative workflows will become clearer.
For example, today's edition would include:
AI-summarized, only the topics you pick — one digest a day via Email, Slack, or Discord.
Free · 30 seconds with Google · unsubscribe anytimeWhat is AIToday? →
Ask AI anything about this article. Q&As are published on this page for other readers too.
ITmedia News published an article explaining why AI-generated images strike consumers as odd or uncomfortable…
OpenAI announced ChatGPT Images 2.5 and a new feature called Sketch, which turns your doodles into AI images
Vision AI is emerging as a site-wide safety layer for construction, analyzing live video from cameras and dron…
The organizer of the Ibusuki Marathon in Kagoshima confirmed on September 8 that generative AI was used to cre…
Adobe introduced Generative Media in Premiere, letting editors highlight an empty gap and generate context-awa…
Restaurants are using generative AI to create food ads instead of photographing real meals
