Image Resizer – Bulk Resize to Many Sizes

Give it image links and the sizes you need. Every image comes back in all of them in one run — thumbnail, web size, exact square crop, padded social banner — each file named after its size, with the exact dimensions you actually got.

Quick start

1. Paste your image links into Image URLs (or upload a text file with one URL per line).
2. Leave the Sizes as they are, or write your own — one per line.
3. Click Start. The Output tab shows every file with its size name and dimensions.

Input example:

{
  "imageUrls": [
    { "url": "https://example.com/photo.jpg" },
    { "url": "https://example.com/logo.png" }
  ],
  "sizes": ["thumb:400x400 crop", "medium:800", "og:1200x630 pad", "large:1600"],
  "format": "webp",
  "zip": true
}

Output row (shortened):

{
  "sourceUrl": "https://example.com/photo.jpg",
  "size": "thumb",
  "requested": "thumb:400x400 crop",
  "status": "resized",
  "fileUrl": "https://api.apify.com/v2/key-value-stores/.../photo-3f1c2a9b7e-thumb.webp",
  "width": 400,
  "height": 400,
  "bytes": 29307,
  "originalWidth": 1280,
  "originalHeight": 1709,
  "format": "webp",
  "cropped": true,
  "upscaled": false,
  "chargedAsImages": 1
}

How to write a size

You write	You get
800	At most 800 px wide, height follows the picture
x600	At most 600 px tall, width follows the picture
800x600	Fits inside 800 × 600 — nothing cut off, so one side is usually smaller
800x600 crop	Exactly 800 × 600, cut from the middle
800x600 smart	Exactly 800 × 600, cut around the busiest part of the picture
1200x630 pad	Exactly 1200 × 630, the gaps filled with your Padding colour
thumb:400x400 crop	The same, and the file is named …-thumb.jpg

Run it after a scraper

On your scraper or its task, open Integrations, search for this Actor and set its input:

{
  "datasetId": "{{resource.defaultDatasetId}}",
  "urlField": "images.url"
}

Every time the scraper finishes, its images go through this Actor. Leave URL field empty and it finds the image links itself; use a dot path (images.0.url, images.url for a whole list) when you want one exact field. Each row keeps recordIndexes, the dataset rows the image came from.

Settings worth knowing

Setting	What it does
Output format	Keep the original, or convert everything to WebP, JPEG, PNG or AVIF as it is resized.
Quality	82 by default, for JPEG, WebP and AVIF. PNG and GIF stay lossless.
Enlarge images that are too small	Off by default. A small source is returned at its own size rather than blown up, and the row says upscaled: false.
Padding colour	Only for pad sizes. #ffffff, #0000 for transparent, or a name like white.
Put everything in one ZIP	One file to download instead of clicking through a list.
Max file size	Skip source images bigger than this instead of spending the run on them.

Why this one

Common problem	What happens here
One run per size, so four runs for four sizes	Write all the sizes once; every image is made in all of them in a single run.
Files all called output-1.jpg	Each file is named after the source and the size, e.g. photo-3f1c2a9b7e-thumb.jpg.
"Is this really 400 × 400?"	Every row carries the dimensions you got, the original dimensions, and whether it was cropped or enlarged.
Small images come back blurry and stretched	Nothing is enlarged unless you ask, and a crop bigger than the source keeps the shape instead of inventing pixels.
Transparency turns black	Alpha is kept, and pad can fill with #0000 to stay transparent.
Animated GIFs lose their animation	Animated sources keep every frame.
Photos come out rotated	EXIF orientation is applied before resizing.
"Out of memory" on big batches	Images are resized a few at a time within the run's memory, and one too large gets a clear error instead of crashing the run.
One site gets hammered and blocks you	At most 4 images at a time per site, with an automatic slow-down if a site rate-limits.
Paying for broken links	404s, pages that are not images and oversized files are never charged.

Pricing

Pay per file produced. Files that fail are never charged, and there is no fee just for starting a run.

A resized file up to 3 megapixels (about 2048 × 1448, which is larger than most things you resize to) counts as one — so a thumbnail from a 40-megapixel photo is still one. Only a file bigger than 3 megapixels counts one more per 3 megapixels. Each row says what it was charged as in chargedAsImages.

Four sizes of one image are four files, so one source image at four sizes counts as four.

Call it from your code

curl -X POST "https://api.apify.com/v2/acts/tidyrun~image-resizer/run-sync-get-dataset-items?token=YOUR_TOKEN" \
  -H 'Content-Type: application/json' \
  -d '{"imageUrls":[{"url":"https://example.com/photo.jpg"}],"sizes":["thumb:400x400 crop","large:1600"]}'

Tips

• Building a responsive srcset? 400, 800, 1600 in one run gives you the three files and their exact widths for the w descriptors.
• Product grid? thumb:400x400 crop keeps every tile the same shape; pad with #ffffff if you would rather not cut the product.
• Social preview? 1200x630 pad is the usual Open Graph size, and padding keeps the whole picture visible.
• Want them smaller as well as shorter? Set Output format to WebP — resizing and converting happen in the same pass.

Responsible use

Downloads the image URLs you provide, as an ordinary visitor. It does not log in, does not solve CAPTCHAs and does not collect personal data. Only resize images you have the right to use.