- Free allowance
- 79,657 free eBooks, 'No fees, no registration, completely free'.
- Verdict
- caveatusable, with a named limit or an unprobed read path
- Cost
- free tier
- Identity
- anonymousno card, no phone
- Binding limit
- per-ip-shapingautomated access from the main site is blocked by IP; 'Such blocks automatically expire after a few days'
- Evidence
- documented5 sources; a first-party page was read and the sentence quoted
- Freshness
- stale on 2026-11-23checked 2026-10-23 (live); stale on 2026-11-23
| Free tier | 79,657 free eBooks, 'No fees, no registration, completely free'. The published collection metadata gives the real size: main collection 3,443,023 files / 1,064,736,252,276 bytes, generated collection (EPUB, MOBI, HTML) 2,681,516 files / 1,375,569,708,792 bytes, i.e. 6,124,539 files and 2,440,305,961,068 bytes (2.44 TB), metadata updated 2026-10-19. unit TB; metered on downloaded-bytes; The 79,657 counts titles; metadata.json counts files. Both were read on the same day and they describe different things, so neither is the other's check. |
|---|
| Price | USD 0 |
|---|
| Limits | - per-ip-shaping (binding) automated access from the main site is blocked by IP; 'Such blocks automatically expire after a few days' source
- request-rate the vendor's own harvest example uses wget -w 2 (two seconds between requests); mirrors are the approved bulk path source
|
|---|
| Binding | per-ip-shaping At 1 TiB/month the main site blocks the caller long before any byte cap; the mirror network is the only path that carries that volume, and the mirror list is what makes the quota effectively unlimited. |
|---|
| Identity | anonymous The home page states 'No fees, no registration, completely free'. No account exists to create. |
|---|
| Egress | Free and unmetered from the main site and from every mirror. The main site is not a bulk endpoint: 'This website is intended for human users only. Any perceived use of automated tools to access this website will result in a temporary or permanent block of your IP address.' Bulk is approved through rsync mirrors, which the mirroring guide documents. free allowance: no byte allowance published |
|---|
| Per-object cap | not applicable: one file per book per format. No per-file cap is published; the largest objects in the tree are audio ZIPs, and the file list is the only inventory. |
|---|
| Retention | permanent: the mirroring guide describes a fixed collection that 'changes daily, as new items are added or updated', with no expiry clock. |
|---|
| Interface | rest, rsync, cli plain HTTPS per file (https://www.gutenberg.org/cache/epub/<id>/pg<id>.txt), an OPDS feed at /ebooks/search.opds/, a JSON OPDS2 feed 'currently available for testing', GUTINDEX.ALL and metadata.json for inventory, and rsync mirrors listed in MIRRORS.ALL. |
|---|
| Tooling | wget/curl, rsync against any mirror, the OPDS feed for readers. The robot policy's own example is 'wget -w 2 -m -H "https://www.gutenberg.org/robot/harvest?filetypes[]=html"' - two seconds between requests, from the main site, for a bounded harvest. |
|---|
| Composes with | The canonical public-domain text corpus to pair with LibriVox audio and Internet Archive scans of the same titles; a 2.44 TB rsync mirror can be filtered to the zip files only, which drops the audio weight. |
|---|
| Reliability | Main-site IP blocks are automatic and last days; the generated collection is rebuilt in batches, so a mirror refresh sees traffic spikes rather than a steady delta. |
|---|
| Notes | Caveat is the main site, not the catalogue. The abandoned route is 'scrape gutenberg.org with parallel connections': the robot policy names it as a blockable offence and the mirror path costs nothing. |
|---|
| Evidence note | documented The collection size is the vendor's own metadata.json, fetched here; one complete file was downloaded and hashed. What would raise it to measured for the bulk path is an rsync of a filtered subset from a mirror with a byte count, which was not run. |
|---|
| Parsed numbers | - has-free-tier 1
- price USD 0
|
|---|