Skip to content
ImageQuarry

Question

What is instruction-based image editing?

Instruction-based image editing means changing an image by writing the change, such as "replace the background with a beach", instead of using editing tools. The model reads the image and the instruction together and returns an edited image.

It differs from classic inpainting, where you paint a mask over the area to change. Google describes defining that area in conversation instead, aiming to leave the rest of the image as it is.

OpenAI's image API supports edits from a prompt and edits guided by a mask, and it notes that masking with its model is prompt-based: the mask is guidance, not an exact boundary.

The trade-off: it is less exact than manual work for fine pixel changes, and areas you did not mention can shift, so compare the result with the original.

ImageQuarry is an instruction-based workspace: create an image, edit it with words, keep refining.

Sources: Google's developer documentation, Gemini image generation; OpenAI image generation guide (masks); checked October 20, 2026.

Next step

Can you edit a photo by describing the change in words?Whether it works on your own photo.Read more

Related questions

Can you edit a photo by describing the change in words?Yes. Write the change and the model applies it.Read more What is inpainting?Filling or replacing a chosen area of an image.Read more What is a mask in AI image editing?A marked area that tells the model where to edit.Read more What is the difference between an AI image generator and an AI image editor?One makes new images, the other changes existing ones.Read more

Last reviewed October 20, 2026 by ImageQuarry

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

Edit with words

Create an image, describe the next change, and keep refining.

Start creatingAlready have an account? Sign in