GPT-image-1.5 released with enhanced editing and face preservation — limited access

AI Impact Summary

GPT-image-1.5 is now available through Azure OpenAI, building on GPT-image-1 with higher quality image generation, editing controls, and face preservation. It supports text-to-image and image-to-image editing (inpainting) with high input fidelity, enabling editing of a single aspect while preserving others, and outputs up to 1024x1536 or 1536x1024. Access is limited via a model application process, so teams should prepare onboarding plans, refactor workflows to leverage editing-mode capabilities, and consider governance around face-preservation use.