Precise image editing that changes only what's asked, keeping subject, composition, and background intact, with reference subjects staying recognizable across styles and successive edits.
Scores appear here when an evaluation matches this exact model.
Editing built for the tightest control, edits scoped precisely to the instruction, with subject and composition preserved across many rounds of revision.
General-purpose GPT Image 2.5 tier with transparent output, custom dimensions and five explicit quality levels.
Trades longer generation time for intricate detail; supports transparent backgrounds and quality levels through max.
OpenAI's GPT Image 1 generates and edits images via the dedicated Images API. Features accurate text rendering, transparent backgrounds, and up to 16 reference images for edits.