Gpt-image-1
Gpt-image-1
Version: 2025-04-15
OpenAILast updated April 2025
An efficient AI solution for diverse text and image tasks, including text to image, image to image, inpainting, and prompt transformation.
Multimodal
Gpt-image-1 includes significant improvements on image generation: it has generate high-quality images in challenging scenarios and solving challenging prompts.
Currently the model has great zero-shot capabilities in image in-painting,
photorealistic photos, and wireframe designs.

Key Capabilities

  • More modalities Gpt-image-1 includes both text to image and image to image modalities
  • Inpainting: Edit images with text prompt and user drawn bounding box
  • Prompt transformation: Edit images with text prompt
Model Specifications
LicenseCustom
Last UpdatedApril 2025
Input TypeText,Image
Output TypeText,Image
PublisherOpenAI
Languages1 Language