Image-to-image
Image-to-image is the task of generating a new image from an existing one plus a prompt, keeping some of the original's structure while changing style, detail or content.
Also called: img2img, image editing, image transformation
By Divya Sharma · Checked Nov 15, 2026 · How we review
Instead of starting from pure noise as in text-to-image, the model starts from your picture. You might turn a rough sketch into a finished illustration, restyle a photo as a painting, or change the season in a landscape.
How it works
The common img2img method comes from SDEdit. It adds a chosen amount of noise to the input image, then lets a diffusion model denoise it again. The strength or denoise setting decides how much noise is added. Low strength keeps most of the original; high strength keeps only rough shapes and colours and gives the prompt more say.
Instruction editing
Newer models accept a plain instruction such as "make the jacket blue" and edit the image directly, aiming to change only what was asked. Qwen Image Edit 2511 is one open example. These are also called image editing models.
Related terms
Inpainting edits only a masked area. ControlNet keeps a specific structure, such as edges, depth or pose, rather than the whole image. Upscaling with an AI upscaler is another image-to-image task, focused on resolution.
Limits
There is a trade-off between staying faithful to the input and following the prompt. Faces and small text can change in unwanted ways, even at low strength.
Related terms
- Inpainting and outpainting
- Inpainting is filling a masked region of an image with new generated content that fits the rest of the picture. Outpainting is the same idea applied outside the edges, extending the image beyond its frame.
- Text-to-image
- Text-to-image is the task of generating a new picture from a written description, called a prompt. Most current text-to-image models are diffusion models conditioned on the text.
- ControlNet
- ControlNet is a neural network add-on that lets a text-to-image diffusion model follow a structural guide, such as an edge map, depth map or human pose, in addition to the text prompt.
- Diffusion model
- A diffusion model is a generative model that learns to turn random noise into data, such as an image, by removing a little noise at a time over many steps.
- AI image upscaling
- AI image upscaling, or super-resolution, uses a neural network to enlarge an image and fill in plausible detail, so the result looks sharper than simple resizing would make it.