Overview
- OpenAI says the update is rolling out globally to all ChatGPT users and is available via the API as GPT Image 1.5, with Business and Enterprise workspace access coming later.
- The company claims image generation is up to four times faster and that edits now preserve lighting, composition and facial likeness while changing only what users request.
- New capabilities include targeted actions to add, remove, blend, combine and transpose elements, along with stronger instruction following for complex, multi‑step prompts.
- ChatGPT gains a sidebar Images section with presets, trending prompts and an image library that supports easier iteration and parallel image generation.
- OpenAI published public demos and prompts, including a Sam Altman K‑pop example, as independent comparisons pit the tool against Google’s Nano Banana Pro with some tests crediting Google on photorealism.