Help Center
English
Contents

Edit and crop images

Change an image with AI, reframe it, remove its background, or crop it, and undo a replace.

Open an image in Media, Site assets or Ad assets to edit it with AI, remove its background or crop it. All need edit access to that library and storage space; editing with AI and removing the background also need access to image generation and available AI credits.

JPEG, PNG, WebP and GIF images can be edited and cropped. Animated images cannot be cropped.

Keep the original or replace it

Before a crop, an AI edit, a reframe or a background removal starts, choose where the result goes. New copy saves it as a new file in the same library and leaves the original as it is, so products, pages and ads keep showing the original until you choose the new file. Replace original puts the result in place of the original's image, with the same link, everywhere the image is used. Before anything changes, the window lists where the image is used (products and variants, collections, categories, brands and vendors, blog posts, pages, policies, emails, theme settings such as the logo and favicon, social sharing images and orders) and asks you to confirm.

Undo a replace and earlier versions

A replaced image is never lost: it is kept as an earlier version. Right after a replace, choose Undo in the message to put it back. Later, open the image's details: Earlier versions lists the images it replaced, newest first, with Restore and Delete. Restore puts one back everywhere the image is used and keeps the current image as an earlier version, so a restore can be undone too. The newest 5 versions are kept; the oldest goes first. Earlier versions count toward your storage and are deleted with the file. The storefront shows the new or restored image within about 30 seconds.

Edit an image with AI
  1. Open the image's details in its library.

  2. Choose Edit with AI.

  3. Describe the change in your own words, for example "put it on a marble table" or "make it warmer", or start from one of the ideas.

  4. Under Background, choose Standard for a JPEG or Transparent for a PNG with a transparent background, then choose the image quality.

  5. Under Save the result as, choose New copy or Replace original. Review the credit cost on the Edit button.

  6. Choose Edit and wait for the result.

The result keeps about the shape of the original.

An edit costs the same as generating an image: 3 AI credits for a standard result, or 6 for HD; a transparent result costs 5 AI credits, or 20 for HD. The Edit button shows the cost before you start. An instruction the image safety policy rejects uses no credits.

Remove the background
  1. Open the image's details in its library.

  2. Choose Remove background.

  3. Choose the image quality and, under Save the result as, New copy or Replace original. Review the credit cost on the button.

  4. Choose Remove background and wait for the result.

The subject is kept on a transparent background and saved as a PNG. It costs the same as a transparent generated image.

Crop and rotate an image
  1. Open the image's details in its library.

  2. Choose Crop.

  3. Choose an aspect ratio: Free, Original, 1:1, 4:5, 9:16 or 16:9. Drag the crop to move it and its corners to resize it, zoom in for detail, and rotate a quarter turn left or right.

  4. Check the size in pixels at the bottom of the window, then choose Save as new file, or Replace original to put the crop in place of the image everywhere it's used.

On a keyboard, the arrow keys move the crop by 1 pixel and Shift with an arrow by 10. On a phone or tablet, drag with one finger.

The crop is cut from the full-resolution original, not from the preview on screen, so it keeps every pixel of detail. A JPEG stays a JPEG; a PNG, a GIF or a WebP with transparency is saved as a PNG and keeps its transparency. The new file keeps the original's alt text and counts toward storage and the file limit like any upload.

With your AI assistant or the Store API
Reframe an image

Reframe puts a photo in a new frame, such as 4:5 for a product page or 16:9 for a banner, and lets you place it there: drag the photo to move it, and zoom it smaller to pull the camera back or larger to move closer. AI fills the empty space by continuing the scene seamlessly, and anything outside the frame is cut. Moving closer is redrawn by AI with real detail, not a blurry enlargement. The product is kept whole: Tringify finds the product in the photo, shows its outline and warns you before a placement would cut it. The new area is made by AI, so check the result before you use it.

  1. Open the image's details in its library and choose Reframe.

  2. Choose the frame. Choose a quick placement, Extend down, Extend up, Extend left, Extend right or Centred, or drag and zoom the photo yourself. For example, Extend down keeps the photo at the top and extends the bottom; dragging the photo up cuts the top and leaves more of the bottom to fill.

  3. Choose the image quality and, under Save the result as, New copy or Replace original. Review the credit cost on the Reframe button; it costs the same as an AI edit.

  4. Choose Reframe and wait for the result.

To reframe several images, select them and choose Reframe in the selection bar, then choose one frame and one placement, for example Extend down to 4:5. Every photo is kept whole and each finished image costs the same as an AI edit.

Make a new image from your images

Select up to 4 images in a library and choose New image from these. Describe the image you want; the model is shown your images beside the description, for example a product photo, a logo and a style to follow. The result is a new file and the images you selected are not changed. It costs the same as any generated image.

A connected AI assistant can reframe images (estimate_image_reframe, then reframe_image) and pass reference_file_ids to generate_image. The Store API offers POST /files/{id}/reframe with aspect_ratio and either direction or zoom, offset_x and offset_y, and reference_file_ids on POST /files/generate.

Turn an image into an animated GIF

Turn into GIF animates an image as a short loop with AI and saves it as a new GIF file; the original stays as it is. Use it for a product photo with a little life in it, such as steam, light or fabric moving.

  1. Open the image's details in its library and choose Turn into GIF, or select several images and choose Turn into GIF in the selection bar.

  2. Describe how it should move if you want to guide it, for example "steam rises from the cup". You can leave this empty.

  3. Choose the length (the number of frames) and the speed. The button shows the credit cost: each frame costs credits, so a longer GIF costs more.

  4. With 2 to 4 images selected, you can choose One GIF from all selected images to make a single animation that moves through them.

  5. Choose the button to start. A GIF can take a few minutes; the bar at the top of the library shows progress and you can leave the page.

A connected AI assistant makes GIFs with create_image_job and the action animate (estimate_image_job first); the Store API offers the same on POST /files/image-jobs.

Make a GIF from several images

Make a GIF joins several images into one looping GIF that shows them one after another, for example the colours of a product or the steps of a recipe. It uses no AI and costs no credits.

  1. Select 2 to 20 images in a library, in the order you want them shown, then choose Make a GIF in the selection bar.

  2. Choose how long each image is shown, then choose the button to start.

The GIF takes the shape of the first image and is at most 640 pixels on its longer edge. Every image is shown whole, centred on white, never cropped. The GIF is saved as a new file; the images you selected are not changed.

A connected AI assistant makes the same GIF with create_image_job and the action slideshow; the Store API offers it on POST /files/image-jobs.

Make product shots
  1. Select one or more product images (up to 25) in Media, Site assets or Ad assets and choose Product shots in the selection bar, or open one image and choose Product shots.

  2. Choose up to 4 scenes: White background, Flat lay, Held in hand, On a table, your own words, or ideas from Suggest with AI. Suggest with AI is free, looks at the first image and can be used 20 times an hour.

  3. Choose the image quality and where the shots are saved: Ad assets or the same library.

  4. Before you start, the window shows how many shots will be made (images × scenes), any images that are skipped and why, the total credit cost, and how many of the store's 300 shots for the last 24 hours are left. One job makes up to 100 shots.

Each shot keeps the product exactly as it is, with the same shape, label and colours, and places it in the scene. Each finished shot is a new file and costs the same as an AI edit; a shot that fails costs nothing. A connected AI assistant or the Store API makes product shots with create_image_job and the action product_shot (estimate_image_job first), and suggests scenes on POST /files/{id}/product-shot-scenes.

On a product's page, Product shots sits on the Product Media card: choose which of the product's images to shoot (the main one is chosen first), and Suggest with AI also reads the product's name. Turn on Also add the shots to this product's images to add each finished shot after the product's images (saved in Media; the main image stays as it is). This needs permission to edit products, the window shows how many more images fit on the product, and unsaved changes to the product must be saved first.

Which images each action works on

Remove background, Edit with AI, Reframe, Crop and New image from these work on still JPEG, PNG, WebP and GIF images. Turn into GIF works on still JPEG, PNG and WebP images; a GIF already is one. Make a GIF works on every image; an animated image shows its first frame.

An image's details show only the actions that image can take. In the selection bar, an action shows when at least one selected image can take it, and the window says how many are skipped and why. Older GIF and WebP images are checked for animation once; until then they work only with Make a GIF.

Work on many images at once

Select up to 100 images in Media, Site assets or Ad assets, then choose Remove background, Edit with AI, Reframe or Crop in the selection bar. The work runs in the background: you can leave the page, and a bar at the top of the library shows progress and a Stop button. A store runs one of these at a time.

  1. Select the images in the library.

  2. Choose Remove background or Edit with AI. The window shows how many of the selected files are images it can work on and the total credit cost. Files that are not images are skipped and cost nothing.

  3. For Edit with AI, describe the one change to make to every image.

  4. Choose the image quality and, under Save the results as, New copies or Replace originals.

  5. Choose the button with the credit cost to start.

In Media or Site assets, choose Copies in Ad assets under Save the results as to put the new files straight into Ad assets, ready for your ads; the originals stay where they are. Turn into GIF and Make a GIF offer the same choice. When the work is done, Open Ad assets in the bar takes you there. This needs permission to manage marketing. A connected AI assistant or the Store API does the same with save_to set to ad_asset on create_image_job or POST /files/image-jobs.

Credits are charged per finished image, never for one that fails or is skipped. If credits run out part way, the work stops and the finished images are kept. Stop ends the work after the images already in progress.

Crop many images to one shape
  1. Select the images, then choose Crop.

  2. Choose the shape every image is cropped to: Original, 1:1, 4:5, 9:16 or 16:9.

  3. Frame each image in turn: drag the crop to move it and its corners to resize it. Use Next image and Previous image to move between them.

  4. When every image is framed, choose the Crop button at the bottom of the window. Every crop is saved as a new file and the originals stay as they are. Cropping costs no credits.

A connected AI assistant can do the same with estimate_image_job, create_image_job, get_image_job and cancel_image_job. The Store API offers POST /files/image-jobs/estimate, POST /files/image-jobs, GET /files/image-jobs/{id} and POST /files/image-jobs/{id}/cancel.

A connected AI assistant can edit images (estimate_image_edit, then edit_image), remove backgrounds (estimate_background_removal, then remove_background), at the same credit cost, and crop them (crop_image). Before replacing an image it checks where it's used (get_file_usage), and it can list and restore earlier versions (list_file_versions, restore_file_version). The Store API offers the same as POST /files/{id}/edit, POST /files/{id}/remove-background and POST /files/{id}/crop, where save_as chooses New copy or Replace original, and GET /files/{id}/usage, GET /files/{id}/versions and POST /files/{id}/versions/{version_id}/restore. See Create images and alt text with your AI assistant.

Was this page helpful?