Bulk Image Downloader

Bulk Image Downloader saves every image from a list of image links, from web pages or from an Apify dataset as real files: each one gets a public download link, and the whole run can come as one ZIP. On the way it can resize, convert to WebP, JPG or PNG, rename files by a pattern, skip duplicates by SHA-256 and drop images below a minimum size. Every row tells you the width, height, bytes, format and hash of the saved file. No login, no API key, no proxy setup.

You pay only for images that were saved. Failed downloads, duplicates, filtered-out images and web pages are free.

At a glance

• Three sources in one run: direct image links, web pages (all images in the HTML), and any Apify dataset with image links
• Files, not just links: each image in the key-value store with a public link, and optionally one images.zip with a metadata.csv
• Resize and convert: maximum width and height, WebP / JPG / PNG, quality setting
• Clean output: file name pattern, SHA-256 dedupe, minimum width, height and file size, format filter
• Reads the real file: format and dimensions come from the bytes, not from the link (JPG, PNG, WebP, GIF, AVIF, HEIC, BMP, TIFF, ICO, SVG)
• Safe by default: size limit per file, private and local network addresses are refused
• Error rows: free, with an errorCode

One real row

Default input, the second link (collected on 2026-10-03):

{
  "url": "https://upload.wikimedia.org/wikipedia/commons/4/47/PNG_transparency_demonstration_1.png",
  "status": "downloaded",
  "pageUrl": null,
  "fileUrl": "https://api.apify.com/v2/key-value-stores/<storeId>/records/000002-PNG_transparency_demonstration_1.png?signature=<signature>",
  "key": "000002-PNG_transparency_demonstration_1.png",
  "fileName": "000002-PNG_transparency_demonstration_1.png",
  "zipFile": null,
  "format": "png",
  "contentType": "image/png",
  "width": 800,
  "height": 600,
  "bytes": 224566,
  "hash": "c4e870d358d64ce677e958c156d0e3fba466993d8e915fc81c5dc64937a9b696",
  "converted": false,
  "scrapedAt": "2026-10-03T13:20:17Z"
}

An image found on a web page also carries pageUrl, the alt text and where it was found (foundIn: img, meta, background, poster, link).

Measured runs

Cloud runs on 2026-10-03, default memory (1 GB):

Run	Result	Time
240 image links, files + one ZIP	230 saved (26 MB), 10 dead links listed as NOT_FOUND, free	24 s
Dataset of a social media scraper (150 items, image field detected automatically), convert to JPG at 1,200 px, ZIP	150 saved	27 s
Dataset of a product scraper, field images, minimum width 300 px, ZIP only	300 saved, 17 duplicates skipped	26 s
2 web pages, minimum width 100 px, ZIP	47 saved, 16 small images filtered out	10 s
5 large photos (5 to 660 MB) with the default 25 MB limit	4 saved (41 MB), the 660 MB file refused in under a second	11 s

What you can do with it

• AI and computer-vision datasets: turn a list of links into same-format, same-size files with a hash per file and no duplicates.
• Product catalogs: point it at the dataset of your e-commerce scraper and get every product photo as WebP at 1,200 px, named your way.
• Website migration and backup: save every image of a list of pages before a redesign, with the page each one came from.
• Content and SEO audits: find oversized or broken images: the row has bytes, dimensions, format and an error code when the link is dead.

Input

Field	What it does
urls	Image links and web page links, mixed. A page is opened and all images in its HTML are downloaded
datasetId	An Apify dataset of your account (pick it in the form or paste its ID); image links are taken from every item
datasetFields	Fields that hold the links (images, product.imageUrl). Empty = detect automatically
maxImages	Stop after this many saved images
maxImagesPerPage	Image links taken from each page (default 500)
createZip	One images.zip with all files and a metadata.csv
saveToKeyValueStore	Each image as its own file with a public link (default on)
fileNamePattern	{index}, {name}, {hash}, {host}, {page}, {width}, {height}, {ext}. Default {index}-{name}.{ext}
convertTo	original, jpg, png or webp
resizeMaxWidth, resizeMaxHeight	Shrink larger images, keeping the aspect ratio. Never enlarges
quality	JPG / WebP quality, default 85
minWidth, minHeight, minFileSizeKB	Skip small images (icons, tracking pixels). Free
maxFileSizeMB	Larger files are not downloaded (default 25, up to 200)
formats	Keep only these real formats, e.g. jpg, png, webp
dedupeByHash	Same file from different links is saved once (default on)
urlMustMatch, urlMustNotMatch	Regex filters for image links found on pages or in the dataset
includeMetaImages, includeBackgroundImages, includeLinkedImages	What counts as an image on a page: social preview, CSS background in style, full-size files linked from thumbnails
maxConcurrency, downloadTimeoutSecs, proxyConfiguration	Advanced

Download a list of image links as WebP thumbnails

{
  "urls": [
    "https://picsum.photos/id/237/1200/800",
    "https://upload.wikimedia.org/wikipedia/commons/4/47/PNG_transparency_demonstration_1.png",
    "https://images.unsplash.com/photo-1506744038136-46273834b3fb?w=1600"
  ],
  "convertTo": "webp",
  "resizeMaxWidth": 512,
  "resizeMaxHeight": 512,
  "fileNamePattern": "{hash}.{ext}"
}

Download all images from a web page to a ZIP

{
  "urls": ["https://books.toscrape.com/"],
  "createZip": true,
  "minWidth": 100,
  "fileNamePattern": "{page}-{index}.{ext}"
}

Download the images of a dataset

{
  "datasetId": "<dataset ID of a finished run>",
  "datasetFields": ["images"],
  "convertTo": "jpg",
  "resizeMaxWidth": 1200,
  "createZip": true
}

Output

One row per image link that was tried.

Field	Description
url	The source link
status	downloaded, duplicate, filtered or failed
fileUrl	Public link of the saved file in the key-value store
fileName, key	Name of the file (also its key in the key-value store and its name inside the ZIP)
zipFile	ZIP part that holds the file (images.zip, images-part-2.zip...)
format, contentType	Real format of the saved file
width, height, bytes	Of the saved file (after resize or conversion)
hash	SHA-256 of the source file
converted, originalFormat, originalWidth, originalHeight, originalBytes	When the file was resized or converted
pageUrl, alt, foundIn	For images found on a web page
sourceField	For images that came from a dataset
finalUrl	When the link redirected
animated, warning, extraMegabytes	Animated file; why a conversion was skipped; MB above the first 5
errorCode, error	Why the image was not saved (never charged)

Where the files are: the Output tab has the links to the files and to images.zip; the Storage > Key-value store tab lists every file. The record SUMMARY has the totals of the run.

Error codes (never charged)

errorCode	Meaning
NOT_FOUND	The server answered 404 or 410
FORBIDDEN	The site refused the download even with a referer and a proxy
HTTP_ERROR, TIMEOUT, DOWNLOAD_FAILED	The server failed or did not answer after retries
NOT_AN_IMAGE, EMPTY_FILE	The link returned something that is not an image
TOO_LARGE	Above maxFileSizeMB, or more than 160 megapixels
TOO_SMALL, FORMAT_FILTERED	Removed by your filters (status: filtered)
DUPLICATE	Same content as an image already saved in this run
NO_IMAGES_FOUND	A page or dataset without image links
INVALID_URL, HOST_NOT_FOUND, BLOCKED_HOST	Not an http(s) link, unknown host, or a private / local network address
DATASET_NOT_FOUND	The dataset ID does not exist in your account
NOT_REACHED	The run timeout or your maximum charge arrived; the list of pending links is in the record NOT_PROCESSED

Pricing

Pay per event, no monthly fee, no start fee, and no Apify compute bill on top.

Event	Free plan	Bronze	Silver	Gold and above
Image downloaded (per 1,000)	US$ 45	US$ 6.50	US$ 5.50	US$ 4.50
Extra MB of a large image	US$ 0.001	US$ 0.001	US$ 0.001	US$ 0.001

• Resize, conversion, public link and ZIP are included in the image price.
• The image price covers source files up to 5 MB, which is almost every image on the web. Only larger files pay the extra MB: a 7.2 MB photo is 3 extra MB (US$ 0.003).
• Failed downloads, duplicates, filtered-out images, web pages and links that are not images are free.
• On the Apify Free plan a run saves up to 100 images.

How to use it from code, on a schedule and in integrations

• API: POST https://api.apify.com/v2/acts/scrapewise~bulk-image-downloader/run-sync-get-dataset-items?token=<YOUR_TOKEN> with the input JSON returns the rows when the run ends. Each fileUrl is a direct download link.
• After another Actor: add this Actor as an integration of your scraper and pass {{resource.defaultDatasetId}} as datasetId: every finished scrape gets its images downloaded. The Actor runs with limited permissions: it can read only the dataset you give it.
• Schedule: save the input as a task and schedule it in the Apify Console.
• n8n, Make, Zapier: use the Apify app, run this Actor and read the dataset items.

FAQ

How do I download all images at once? Turn on createZip and open the ZIP link in the Output tab (also shown in the status message of the run).

How long are the files kept? As long as the run's storage is kept by your Apify plan (7 days for unnamed storages on most plans). Download the ZIP or copy the files before that.

Does it open pages with a browser? No. It reads the HTML the server sends, including lazy-load attributes, srcset (the largest version), social preview tags and inline CSS backgrounds. Images that only appear after JavaScript runs are not seen.

Which image is taken from srcset? The largest one.

Are my files public? The link works for anyone who has it, as with every Apify key-value store record. Nobody can list your store without your token.

Can it keep the folder structure? File names are flat. Use {page} or {host} in fileNamePattern to group files by source.

Limitations

• No JavaScript rendering and no crawling: only the pages you list are read, one level.
• Animated GIF/WebP and SVG are saved as they are (no resize or conversion). Images above 80 megapixels are saved without conversion.
• Files above 4 MB are downloaded one at a time to keep memory low. For files above 50 MB, raise maxFileSizeMB and give the run 2 GB of memory or more.
• Resized or converted files lose EXIF metadata (the orientation is applied first).
• Sites that block datacenter servers may answer FORBIDDEN; choose a proxy in proxyConfiguration for those.
• Only download images you have the right to use. You are responsible for the copyright and the terms of the sources. This Actor reads public links only.
• Something broke? Open an issue on the Issues tab with the run ID; it is answered within hours.

Changelog

• 2026-10-03, 0.1: first release. Image links, web pages and datasets; key-value store files with public links; ZIP with metadata.csv; resize; WebP/JPG/PNG conversion; file name pattern; SHA-256 dedupe; size and format filters; free error rows.