AI Image to Image Editor
Upload a photo, say what should change and keep everything else — same face, new light, new scene. From 2 credits per image depending on the model, nothing if it fails. Use it to make the matching last frame for a transition.
- per image
- ~20s
- per image
- from 2 credits per image
- 18 models
- from 2 credits per image
- if it fails
- 0 credits
- if it fails
- max upload
- 10 MB
- max upload
Sign in to start — images from 2 credits, no card needed to sign up.
Example images
Tap one to start from it: describe a change and it keeps everything else.
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.
Film look
Turn this into a cinematic film still at golden hour, warm backlight, keep the face and pose exactly the same
Next moment
The same scene a few seconds later: the camera has pushed in and the person has turned to face us
New background
Replace the background with a quiet Tokyo street at night, keep the person exactly as is
Image to Image specifications
- Input
- One photo (JPG, PNG or WebP, up to 10 MB) plus a sentence describing the change
- Model
- Nano Banana 2 (Google), strong at keeping faces and scenes consistent
- Sizes
- Same as the input by default, or pick 16:9, 9:16, 1:1, 4:3, 3:4 or 21:9
- Cost
- From 2 credits per image, depending on the model you pick. Failed or blocked images cost nothing
- Rules
- Only photos you have rights to, and consent from anyone recognisable. No undressing or sexualising real people
- Output
- JPEG, saved to your Assets and selectable in every image slot
How to use Image to Image
Step 1
Add your photo
Upload one, or pick one you made earlier from your Assets.
Step 2
Say what changes
Describe only the change and name what must stay (“keep the face the same”). Short and specific works best.
Step 3
Make it a video
Use the result as a first frame in Image to Video, or as the “To” frame in the Transition Maker.





