Internet Archive Scraper

Search the Internet Archive and export clean, structured data about its items — books, movies, audio, software, images and web captures — with direct download links. No login, no API key, no browser required.

Every result comes back as a clean, normalized record — archive.org's irregular fields flattened into a stable shape, with a direct, permanent download URL for each file — so it drops straight into your database, spreadsheet or ML pipeline.

What you can do

• Search anything on archive.org — a plain keyword or phrase matches item titles, creators and subjects; power users can pass a full Lucene query.
• Filter by media type (texts, movies, audio, software, images, data, web…) and by collection.
• Sort by relevance, most downloaded, newest/oldest, or trending this week.
• Get download links — optionally pull every file in an item (PDF, EPUB, MP3, MP4, ZIP…) with a direct, stable download URL for each.
• Fetch by identifier — already know the items you want? Pass their identifiers and get full metadata straight away.

Input

Field	Description
searchQuery	Keyword or phrase — matches item title, creator & subject — or a full Lucene query (e.g. grateful dead, or title:(python) AND year:[2015 TO 2020]).
mediaType	Restrict to one media type, or search all.
collection	Optional collection identifier (e.g. librivoxaudio, nasa).
sortBy	Most relevant / most downloaded / newest / oldest / trending.
maxItems	Cap on how many items to return (archive.org allows up to 10,000 per search).
includeFiles	Also fetch each item's full file list with download URLs.
identifiers	Fetch specific items directly by identifier (ignores the search query).

Output

Every item is one dataset record:

{
    "identifier": "principleofrelat00eins",
    "title": "The principle of relativity",
    "creators": ["Einstein, Albert"],
    "date": "1920",
    "mediaType": "texts",
    "collections": ["cdl", "americana"],
    "subjects": ["Relativity (Physics)"],
    "downloads": 12873,
    "detailsUrl": "https://archive.org/details/principleofrelat00eins",
    "thumbnailUrl": "https://archive.org/services/img/principleofrelat00eins",
    "files": [
        {
            "name": "principleofrelat00eins.pdf",
            "format": "Text PDF",
            "sizeBytes": 8123456,
            "downloadUrl": "https://archive.org/download/principleofrelat00eins/principleofrelat00eins.pdf"
        }
    ]
}

(files / filesCount are included only when Include file list & download links is on.)

Example use cases

• Build a dataset of public-domain books, films or audio for research or ML training.
• Monitor a collection for newly added items.
• Resolve a list of archive.org identifiers to full metadata and download URLs.
• Find the most-downloaded items on a topic.

Pricing

Pay-per-event — you only pay for what a run returns:

Event	Price
Actor start	$0.005 per run — one charge per GB of memory (so $0.005 at the default 512 MB)
Archive record	$0.001 per item written to your dataset

A run that matches nothing costs only the start event. Example: a search returning 500 items ≈ $0.005 + 500 × $0.001 = $0.505. Use Maximum items to cap the per-record charges exactly, and turning on Include file list & download links adds no extra charge. Prices are always shown in the Apify Console before you run.