Image to Image AI

Example prompts for Image to Image

The same prompts are behind the “Try an example” button in the input box; each click loads the next one.

• Change season

  Make it winter — snow on the ground and rooftops, bare trees, cold overcast light

• Swap background

  Replace the background with a plain studio backdrop, keep the subject exactly as is

Made with these models

Images generated on this site with the same models.

Will Image to Image work on your photo?

Match what is wrong with your image to what this tool can realistically do. Where it cannot help, the row says which tool can.

• Swap a background, recolour, add or remove an object

  Good match

  These are the edits the models handle most reliably, because you can name both the change and what must stay.

• Keep a face identical through the edit

  Partial improvement

  Add “keep the face exactly the same” and prefer Nano Banana, which follows instructions most literally. Check the result at full size.

• Readable text inside the picture

  Partial improvement

  GPT Image 2.5 and Seedream handle lettering best. Put the exact words in quotes and keep them short.

• Fix a blurry, noisy or damaged photo

  Use another tool

  Editing models repaint; they do not repair. For quality problems use the Image Enhancer

Why Image to Image on AI Image Tools

• Edit with a sentence, not a mask

  "Make it winter", "change the shirt to red", "replace the background with a plain studio backdrop". No brushes, no layers.

• Four editing models, one upload

  FLUX.1 Kontext is the fast default. Nano Banana follows instructions most literally and accepts several reference images. Seedream 5 and GPT Image 2.5 push detail and resolution.

• The rest of the photo survives

  The models are told to change only what you asked for. Name what must stay the same and it will.

• Pay per returned image

  Variations run in parallel and you are charged only for the ones that come back. Failed runs cost nothing.

Every image model in one place

Pick any of these from the Model menu above; the price is shown before you generate. Each name links to what that model is good at. Compare every tier.

• Z-Image Turbo

  Alibaba · cheapest and fastest, text-to-image only

  1 credits per image

• Seedream 5 Pro · 1K

  ByteDance · same model as 2K at half the price

  9 credits per image

• GPT Image 2.5 Flare · 2K

  OpenAI · same price as Sunburst, tuned for speed and volume

  13 credits per image

• Qwen Image 3 · 2K

  Alibaba · strong at readable text inside the image

  6 credits per image

• Nano Banana

  Google Gemini 2.5 Flash · fastest, cheapest

  5 credits per image

• Nano Banana 2.1 · 1K

  Google · newest Nano Banana, sharper text, banner ratios

  10 credits per image

• Nano Banana 2.1 · 2K

  Google · same model at 2K

  15 credits per image

• Nano Banana 2.1 · 4K

  Google · for print or heavy cropping

  23 credits per image

• Ideogram V3 · Turbo

  Ideogram · text inside the image, fastest tier

  5 credits per image

• Ideogram V3 · Balanced

  Ideogram · text inside the image, posters and logos

  9 credits per image

• Ideogram V3 · Quality

  Ideogram · text inside the image, highest detail

  13 credits per image

• FLUX.1 Kontext

  Black Forest Labs · fastest, about 5s per image

  5 credits per image

• Seedream 5 Lite

  ByteDance · good at Chinese text and posters

  7 credits per image

• Seedream 5 Pro · 2K

  ByteDance · most photorealistic, highest detail

  18 credits per image

• GPT Image 2 · 2K

  OpenAI · good with text in images, half the price of 2.5

  7 credits per image

• GPT Image 2 · 4K

  OpenAI · for print or heavy cropping

  10 credits per image

• GPT Image 2.5 · 2K

  OpenAI · newest, better instruction following

  13 credits per image

• GPT Image 2.5 · 4K

  OpenAI · newest, highest resolution

  20 credits per image

Image to Image specifications

Input

One JPG, PNG, WebP or HEIC up to 10 MB

Prompt

One or two sentences: what changes, what stays

Models

FLUX.1 Kontext (default), Nano Banana, Seedream 5 Lite / Pro, GPT Image 2 / 2.5

Sizes

Same as input, 1:1, 16:9, 9:16

Batch

1, 2 or 4 variations per run

Output

PNG or JPEG depending on the model, saved to My Creations

Cost

5 to 20 credits per image, depending on the model

Time

About 25 seconds with Kontext; Seedream around 50 seconds, GPT Image about a minute, Nano Banana can take several minutes when busy

Free plan

10 credits on sign-in — 10 images on the default model; free downloads carry a small watermark

How to use Image to Image

1. Step 1

   Upload the photo

   Any JPG, PNG, WebP or HEIC up to 10 MB. Phone photos are fine.

2. Step 2

   Describe the change

   Name the one thing to change and what should stay. "Snow on the ground and rooftops, keep the building and the people exactly as they are."

3. Step 3

   Generate and compare

   Ask for 2 or 4 variations, pick the best, and run it through the upscaler if you need a bigger file.

Get a better result

A few choices before you upload change the output more than any setting on this page.

• Upload the original file

  Screenshots and images re-saved by messaging apps carry compression the model then has to work around. Find the earliest copy you have.

• One change per run

  Stacking five instructions into one prompt is where results go sideways. Each run costs a few credits and takes about half a minute, so chain them.

• Fix quality first, then edit

  A soft or noisy source produces a soft or noisy edit. Clean it up, then come back. Image Enhancer

How to write a prompt for Image to Image

• One change per run

  Stack edits across runs instead of listing five in one prompt; each run is cheap and fast.

• Say what must stay

  "Keep the face, pose and background" is the most useful sentence you can add.

• Describe the result, not the process

  "The wall is painted sage green" works better than "repaint the wall".

• Match the light

  If you add an object, say where the light comes from so it sits in the scene.