Skip to content
ImageQuarry

Question

What is image-to-image AI?

Image-to-image AI makes a new image from an input image plus a text prompt. The input sets the starting point for composition and content, and the prompt says how to change it, so the result can stay close to the original or move far from it.

Hugging Face's Diffusers documentation describes it as text-to-image with "an initial image as a starting point". A strength setting decides how much the output resembles the input: low keeps it close, high gives the model more flexibility.

It reworks the whole picture, unlike inpainting, which changes only a marked area. Current instruction editors grew out of the same idea: you give an image and describe the change.

To start from your own picture rather than a blank page, see image to image AI.

Source: Hugging Face Diffusers image-to-image and inpainting documentation, checked October 20, 2026.

Next step

Can I upload my own photo and edit it with AI?Starting from a picture you already have.Read more

Related questions

Can I upload my own photo and edit it with AI?Yes, most AI editors start from an uploaded photo.Read more What is a reference image in AI image generation?A picture that shows the model what you mean.Read more What is inpainting?Filling a marked area of an image with new content.Read more What is instruction-based image editing?Editing a picture by describing the change in words.Read more

Last reviewed October 20, 2026 by ImageQuarry

Create an image. Edit it with words. Keep refining.

Start from a picture you have

Upload an image, describe how to change it, and compare the result with the original.

Start creatingAlready have an account? Sign in