Bulk Image Downloader - URLs, Pages & Datasets to ZIP

Problem: your scraper gave you thousands of image links (products, listings, posts), but you need the actual files: stored, named by SKU, without duplicates, maybe resized to WebP, and as one ZIP. Doing it by hand or with a script means retries, rate limits, broken links and disk space.

This actor downloads every image for you from three kinds of input, in any mix:

• Image URLs: direct links, one per line.
• Web pages: every image on the page (<img>, lazy data-src, the largest srcset size, og:image, links to image files).
• Any Apify dataset (datasetId or the Integrations tab): image fields are found automatically, also nested lists like images[].url or media.photos[]. File names can come from a field such as sku, asin or title.

Each file is stored in the run's key-value store with a public link (one result row per file), and all files are packed into images.zip.

Measured: 300 images from 15 pages in 8.7 s (about 2,000 images per minute, 100 MB RAM) with 12 parallel downloads.

What you get that other downloaders don't

• Duplicates are free: the same picture under different URLs (CDN sizes, mirrors) is stored once (SHA-256 of the content) and is not charged.
• No per-MB fee: one flat price per stored file. Failed links, skipped icons and duplicates cost nothing.
• Icons and tracking pixels skipped (min width / height, default 50 px).
• Resize + convert: keep the original, or convert to JPG / WebP / PNG with a max width (for example 2400 px JPG -> 800 px WebP: 505 KB -> 79 KB).
• Only-new mode (monitorId): scheduled runs skip every URL stored before, so you pay only for new images.
• Polite and honest: max 6 parallel downloads per host, retries with back-off on 429/5xx, honest User-Agent, no proxies, no logins.

Input example

{
  "pageUrls": ["https://books.toscrape.com/"],
  "urls": ["https://picsum.photos/id/237/800/600"],
  "outputFormat": "webp",
  "maxWidth": 1200,
  "createZip": true,
  "monitorId": "my-shop-images"
}

Download the images of another actor's run (for example a product scraper), named by SKU:

{
  "datasetId": "YOUR_DATASET_ID",
  "imageFields": ["images"],
  "nameField": "sku",
  "fileNaming": "name"
}

Output example (one row per stored file)

{
  "sourceUrl": "https://books.toscrape.com/media/cache/2c/da/2cdad67c44b002e7ead0cc35693c0e8b.jpg",
  "fileUrl": "https://api.apify.com/v2/key-value-stores/STORE_ID/records/2cdad67c44b002e7ead0cc35693c0e8b.jpg",
  "key": "2cdad67c44b002e7ead0cc35693c0e8b.jpg",
  "fileName": "2cdad67c44b002e7ead0cc35693c0e8b.jpg",
  "contentType": "image/jpeg",
  "bytes": 9876,
  "originalBytes": 9876,
  "width": 125,
  "height": 155,
  "format": "jpg",
  "sha256": "4752d0ef411fb1c4aad31d69425d075c219345c386fd94f43ddfbf4495ea0437",
  "name": null,
  "position": null,
  "foundOn": "https://books.toscrape.com/",
  "downloadedAt": "2026-11-06T00:58:14Z"
}

The ZIP link is in the run's OUTPUT record and in the status message: https://api.apify.com/v2/key-value-stores/STORE_ID/records/images.zip (big runs: images-2.zip, images-3.zip ..., part size set by zipPartMb). RUN_SUMMARY lists counts (stored, duplicates, small, failed), failed links with the reason, and lastError.

Use it from the API

curl -X POST "https://api.apify.com/v2/acts/martin10291~bulk-image-downloader/run-sync-get-dataset-items?token=YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"urls": ["https://picsum.photos/id/237/800/600"], "pageUrls": ["https://books.toscrape.com/"]}'

from apify_client import ApifyClient

client = ApifyClient("YOUR_TOKEN")
run = client.actor("martin10291/bulk-image-downloader").call(run_input={
    "pageUrls": ["https://books.toscrape.com/"], "outputFormat": "webp", "maxWidth": 1200})
for f in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(f["fileUrl"], f["width"], f["height"])
print(client.key_value_store(run["defaultKeyValueStoreId"]).get_record("OUTPUT")["value"]["zip"])

Use it as an integration (chain it after another actor)

1. Open your scraper (products, real estate, Google Maps, social posts) in Apify Console -> Integrations tab.
2. Add integration -> Actor -> pick this actor, set Dataset to {{resource.defaultDatasetId}}, optionally nameField (e.g. sku) and monitorId.
3. Every time the scraper finishes, its images are downloaded, deduplicated and zipped automatically.

Schedule it (only new images, daily)

Create a Schedule (daily or hourly) with the same input and a fixed monitorId. Each run downloads only images that were not stored before: a daily snapshot of new product photos, listing photos or page images, without paying twice.

Use it from AI agents (MCP)

The actor works through the Apify MCP server (https://mcp.apify.com), so Claude, ChatGPT, Cursor or any MCP client can call it as a tool: "download all images from this page as WebP and give me the ZIP link".

Pricing

$1.50 per 1,000 stored files (pay per result). Duplicates, skipped icons, non-images and failed links are not charged. No per-MB fee (files up to maxFileMb, default 25 MB).

FAQ

• Which sites work? Any public image link or page reachable without a login. Pages that build images only with JavaScript after load may give fewer images; then use their dataset (from a browser scraper) as input.
• Social media, YouTube, Instagram? Not supported as pages (they need logins/proxies). Give direct image links from your own data instead.
• How long are files kept? As long as the run's storage is kept by your Apify plan. Download the ZIP or move files to your storage.
• Why was a link skipped? See RUN_SUMMARY: failedExamples (HTTP errors, too big), skippedSmall, skippedDuplicate, skippedType.
• Is it legal to download? You are responsible for having the right to use the images. The actor only downloads public files you point it to, with an honest User-Agent.