Bulk File Downloader — URL List to Storage or ZIP

Download any files in bulk from URL lists or the file links on a web page: PDF, ZIP, CSV, XLSX, DOCX, MP3, MP4 and more. Public download links, real type from the bytes, size, SHA-256, type/glob filters, dedupe, size caps, optional ZIP. $2 per 1,000 files + $0.001 per 10 MB.

Download any files in bulk, from a list of URLs or from the links on a web page. Paste direct links to PDFs, ZIP archives, CSV/XLSX exports, Word documents, MP3s, videos or any other file, or give it pages (a reports archive, a dataset index, a podcast page) and it follows every file link it finds: <a href> links, audio/video sources, embeds and objects. Each file is stored in the run's key-value store with a public download link, and its real type is detected from the file bytes (magic numbers: PDF, ZIP, DOCX/XLSX/PPTX, ODF, EPUB, legacy Office, GZIP/7z/RAR/TAR, MP3/WAV/FLAC/OGG, MP4/MOV/WebM, images, SQLite, Parquet, JSON/CSV/XML…), with the server's content type as a fallback. Every row carries the MIME type, size, SHA-256, source URL, page and link text. Filter by file type or URL glob (*.pdf, /reports/**), skip duplicates by content, cap the size per file (default 100 MB) and in total, and optionally get everything as ZIP files. Use it for archiving public documents, collecting datasets for analysis or machine learning, backing up website downloads, mirroring podcast episodes, or as a file-fetching tool for AI agents and automations. Pricing follows the real cost of moving data: $0.002 per file plus $0.001 per started 10 MB, so 1,000 small files cost about $3. Broken links, web pages, filtered types, oversized files and duplicates are never charged.

What data does it return?

Every run outputs clean, schema-validated rows like these:

fileName	extension	mimeType	bytes	detectedBy	url
00001-minimal-document.pdf	pdf	application/pdf	16978	magic	https://raw.githubusercontent.com/py-pdf/sample-files/main/001-trivial/minimal-document.pdf
00002-horse.mp3	mp3	audio/mpeg	28915	magic	https://www.w3schools.com/html/horse.mp3
00003-addresses.csv	csv	text/csv	328	magic	https://people.sc.fsu.edu/~jburkardt/data/csv/addresses.csv

How much does it cost?

Transparent pay-per-event pricing — you only pay for what you get, and the actor always respects your run budget: set a maximum total charge and it stops cleanly at the cap instead of overrunning.

Event	Price	Charged	Worked example
file-downloaded	$0.002	per file saved (type detection, SHA-256, download link and ZIP included)	1,000 small files = $2 + $1 transfer = $3; 100 PDFs of 2 MB = $0.30
transfer-10mb	$0.001	per started 10 MB of each file saved	a 3 MB file = 1 event ($0.001); a 95 MB video = 10 events ($0.01); 1 GB of large files ≈ $0.10
image-result	$0.003	per image saved, only in image mode (page image scanning)	a page with 40 images = $0.12
apify-actor-start	$0.00005	per run per GB of memory (platform start fee)	One run at 1 GB = $0.00005

Input options

Field	Type	Required	Default	Description
mode	string	no	"files"	Any files (default): every file type from a URL list or page links. Images only: scans pages for every image (img, srcset, lazy, CSS backgrounds) and is billed per image — see Pricing.
urls	array	yes	—	Direct file URLs (any type) and/or web pages whose file links should be downloaded — mix freely. A page's static HTML is scanned for links, media sources and embeds. Links added by JavaScript aren't found.
fileTypes	array	no	[]	Only keep files of these types, e.g. pdf, zip, csv, docx, mp3. Matched against the type detected from the file bytes and against the URL extension. On pages, links with these extensions are followed. Empty = every type (page links: common document, archive, data and media extensions).
urlGlobs	array	no	[]	On pages, also follow links matching these globs, e.g. .pdf, /reports/**, https://cdn.example.com/files/. Use for download links without an extension (/download?id=…). * stays inside one folder, ** crosses folders.
maxFileSizeMb	number	no	100	Skip files larger than this (up to 500 MB). Oversized downloads are aborted early and never charged.
maxTotalSizeMb	number	no	2048	Stop downloading once the files saved add up to this many MB (0 = no cap). Keeps transfer charges predictable.
minFileSizeKb	number	no	0	Skip files smaller than this. Skipped files are not charged.
maxFiles	integer	no	1000	Stop after saving this many files in total.
skipDuplicates	boolean	no	true	Skip files whose content (SHA-256) is identical to one already downloaded, even from different URLs. Duplicates are not charged.
createZip	boolean	no	false	Bundle everything into ZIP file(s) of up to 200 MB each. Links are in the run's OUTPUT record. No extra charge.
respectRobotsTxt	boolean	no	true	Skip pages that the site's robots.txt disallows for crawlers (direct file URLs you list are always fetched).

Limits & honest notes

• Pages are scanned as static HTML without a browser, which is fast and cheap. Links added only by JavaScript are not found; pass those file URLs directly instead.
• Pages disallowed by the site's robots.txt are skipped by default (you can turn this off). If a site's robots.txt can't be fetched (HTTP 5xx or a network error, after one retry), all of that site's pages are skipped (RFC 9309), each with a free error row. Direct file URLs you list are always fetched. Login-protected files are not supported.
• Files over the per-file cap (default 100 MB, up to 500 MB) are aborted early and not charged. The total size cap (default 2 GB) stops the run cleanly; give the run 2-4 GB of memory for many large files. ZIP files are split into parts of up to 200 MB.
• Without a type or glob filter, page links are followed only when they end in a common document, archive, data or media extension (images are left out; add jpg or png to File types to include them).
• Download links work for the lifetime of the run's storage (Apify's default retention for your plan).
• Only download files you have the right to use.
• Internal and private network addresses are always refused, including through redirects or DNS tricks.
• Respects your run budget: set a maximum total charge and the actor stops cleanly at the cap.

Related ActorForge actors

• Bulk Image Downloader — every image from a page, srcset and lazy-loaded included
• Word, Excel & PowerPoint to Markdown — turn downloaded DOCX/XLSX/PPTX into Markdown or JSON
• Image Converter & Compressor

Changelog

• 2026-10-19 — Initial release: any file type from URL lists or page links, magic-byte type detection, type/glob filters, SHA-256 dedupe, per-file and total size caps, ZIP export.

---

Built and maintained by ActorForge. Reliability first: every actor in the portfolio is monitored by an automated nightly self-test, and breakages are typically fixed the same day.