A source without an account
Telegram wants an account before it hands over a file; Usenet wants a subscription. The Internet Archive wants neither. From version 0.5.9, a tgarr that has just been installed - nothing signed in, nothing paid for - can already fill its Videos, Music, Library and Gallery, and can already answer Radarr.
The Archive page
In tgarr's menu: Archive. Five shelves - films, audiobooks, music and radio, books and magazines, pictures. With words it searches archive.org; without, a shelf shows what is downloaded most. Get fetches an item straight from archive.org over HTTPS and puts it where it belongs: a film in Videos, an audiobook in Music with its chapters, a book in the Library, pictures in the Gallery.
- One playable set. The Archive keeps most things in many formats at once. tgarr takes one a browser or a reader can open, not all of them.
- Never a program, and nothing that can only be borrowed: only what the Archive gives to anybody for download.
- At most 8 GB at a time.
- Downloads → Internet Archive shows what is arriving, how far it is, and what arrived or did not.
Radarr, with no account of any kind
When Radarr asks tgarr for a film, the Archive is looked up as well, and what it holds under that title and that year is offered beside what Telegram and Usenet have. A grab is fetched by tgarr and handed to Radarr like any other download. For films that are in the public domain - a great deal of what was made before the 1960s, and more than one would guess after - that is a complete answer.
- Films only. Series are not offered to Sonarr from the Archive: its items are not laid out as seasons and episodes reliably enough.
- Trailers, soundtracks and items of another year are left out.
- To switch it off: config key
newznab_archive = false.
What tgarr.me keeps
The Archive holds some 85 million items, and nearly all of them nobody ever downloads. tgarr.me keeps a list of the ones people do - a thousand downloads or more, and every LibriVox audiobook - so that a long Radarr list is answered here and does not put every title to the Archive's own search.
| Shelf | Items listed |
|---|---|
| Books and magazines | 2,388,000 |
| Music and radio | 786,000 |
| Films | 452,000 |
| Pictures | 143,000 |
| Audiobooks | 22,000 |
As listed on 15 November 2026. For each item: its identifier at the Archive, title, maker, year, how often it was downloaded, and its size. That is all. No file, no cover, no text of any book.
Disc images, on a phone
Much of what the Archive has as film is a DVD image, an avi or an old mpeg - downloaded, a file no phone or tablet can open. tgarr converts while it is watched: the watch page plays such a file as H.264 and AAC from whatever second the slider is put on. Nothing is written to the disk and the file is never changed.
- DVD and Blu-ray images (
.iso), avi, wmv, flv, mpg, rmvb, ts. An mkv or mov is tried as it is first. - On a small machine choose 480p. At most three conversions run at a time.
- Not yet: audio CD images, and DVDs that are still encrypted.
What it is not
- Not a mirror. tgarr.me holds names and numbers; the Archive holds the files and decides what it offers.
- Not a way around lending. What the Archive lends for an hour or to a signed-in reader is not taken.
- Not a judgment on rights. "Public domain" differs from country to country; the Archive's page for an item says what is known about it, and what you fetch is yours to answer for.
The Internet Archive is a non-profit library and lives on donations. If tgarr's third source is of use to you, it is the Archive that deserves the thanks.
Questions: [email protected] or GitHub issues.