AI Picture Editor performs text-prompt-driven photo editing: background removal and replacement, object erasure, old-photo restoration, portrait retouching and style application, all described in plain language rather than manipulated with traditional editing tools. It runs on GPT Image 2, accepts up to 16 reference images for image-to-image generation, and exports at up to 4K with no watermark.
Prompt-based editing lowers the skill floor for edits that traditionally require real Photoshop competence – object removal and photo restoration in particular are technically demanding manual tasks that a well-phrased text instruction now handles directly. Supporting up to 16 reference images per generation also allows more complex, multi-source edits than a single before/after image pair.
Prompt-based editing can misinterpret ambiguous instructions in ways a manual tool wouldn’t, so results on anything precise (exact object boundaries, specific color matching) are worth checking rather than assumed correct on the first try. Credits convert to a limited number of edits per tier (roughly 120-800 depending on plan), worth modeling against real editing volume.
