Internet Archive (IAS3 / petabox item upload)
- Free allowance
- 1 TB hard limit per item page, with a recommendation of 1,000 files or 500 GB per item, and a single file should not exceed about 500 GB.
- Verdict
- ★ qualifiesusable as stated; the read path was probed
- Cost
- free tier
- Identity
- emailno card, no phone
- Binding limit
- per-object-cap1 TB/item; 500 GB of data or 1,000 files per item recommended; bulk-upload guidance says 500 files or 500 GB
- Evidence
- documented16 sources; a first-party page was read and the sentence quoted
- Freshness
- stale on 2026-11-21checked 2026-10-21 (inherited); stale on 2026-11-21
| Free tier | 1 TB hard limit per item page, with a recommendation of 1,000 files or 500 GB per item, and a single file should not exceed about 500 GB. unit TB; metered on unstated |
|---|---|
| Price | none stated |
| Limits |
1 TB/item hard, 500 GB/item recommended; 500-700 GB single-file ceiling; 5,000 files/day across all items (zip counts as 1); recommended <=10,000 files/item but API accepts 250,000; DELETE bucket not allowed; S3 Range headers ignored; only HTTP/1.1 REST, no ACLs; every upload enters a per-item catalog queue plus a global derive queue that can block writes (503-slowdown-spam). Use x-archive-queue-derive:0 to skip expensive derivation. |
| Identity | email A free archive.org account asks for an email address, a screen name and a password, with a reCAPTCHA at sign-up; uploads need that account signed in, and IAS3 keys are generated at archive.org/account/s3.php from the same account. |
| Egress | Unmetered and free: every file in every item is a world-readable HTTP URL (https://archive.org/download/<id>/<file>) with no auth and no egress bill; IAS3 doc states "ACLs are fake. permissions are: World readable, Item uploader writable." Every public item additionally gets an _archive.torrent with IA webseeding as a second free distribution path. |
| Per-object cap | 1 TB/item (1000000000000 bytes) |
| Retention | not stated |
| Interface | s3-compatible, rest, rclone, cli part of the text of Limits, above |
| Tooling | IAS3 = S3-compatible REST (PUT = create item, multipart, x-archive-auto-make-bucket, x-archive-queue-derive:0, x-archive-size-hint); rclone backend 'internetarchive' (Tier 3, added v1.59, github.com/rclone/rclone 60.4k stars); official CLI/Python 'ia' github.com/jjjake/internetarchive (1,923 stars, pushed 2026-10-20); Rust client github.com/LucaCappelletti94/internetarchive-rs. S3 keys from archive.org/account/s3.php. |
| Composes with | Unlimited items per account x 1 TB/item = unbounded per single free account. Also unlimited free accounts. Zip many small files ("a zip file is considered one file" for the daily cap). |
| Reliability | Writes are asynchronous through a per-item queue; the global derive queue can stall uploads and deletes for hours during backlogs, and large/high-file-count items time out and fail to derive. Items are permanent (no TTL) but nothing is guaranteed private. |
| Notes | 1 TB hard limit per item page ("Currently, there is a 1TB size limit for item pages"), recommended 500 GB / 1,000 files per item; items per account unlimited. Single file "should not be larger than ~500 to ~700 GB". |
| Evidence note | documented The page states the number in the record; the claim is documented, not measured. |
| Parsed numbers |
|
Evidence 16
| date | source | what it proved |
|---|---|---|
| 2026-10-22 | help.archive.org/help/uploading-tips/ | 1 TB per item, 500 GB recommended. Re-read https://help.archive.org/help/uploading-tips/: HTTP 200, 5433 bytes, sha256 48c2cd2e54f5effa. However, from a systems perspective, we recommend a maximum of 1000 files or 500GB of data per item page. Keep in mind that most data files that are upload are used to derive multiple files that are used for the site and to allow us |
| 2026-10-21 | help.archive.org/help/uploading-tips/ | 'Currently, there is a 1TB size limit for item pages... we recommend a maximum of 1000 files or 500GB of data per item page' |
| 2026-10-21 | web.archive.org/web/20260113031534/https://archive.org/devel | PUT creates item, multipart, x-archive-auto-make-bucket, 'ACLs are fake... World readable', item->bucket mapping, https://s3.us.archive.org |
| 2026-10-21 | archivesupport.zendesk.com/hc/en-us/articles/360016700691-Up | 'single file ... should not be larger than ~500 to ~700 GB'; 'less than 10,000 files, and where the files do not exceed 1TB'; 'API will allow up to 250,000 files' |
| 2026-10-21 | rclone.org/internetarchive/ | rclone backend exists (Tier 3), notes per-item write queue and derive queue backpressure |
| 2026-10-21 | help.archive.org/help/archive-bittorrents/ | 1.4M items / ~1 PB distributed by BitTorrent as extra free egress |
| 2026-10-22 | probe: https://archive.org/download/nasa/__ia_thumb.jpg | ranged GET 2026-10-22: 302 to a storage node then 206 Partial Content, 1,024 bytes, Content-Range bytes 0-1023/5399; anonymous |
| 2026-11-01 | help.archive.org/help/uploading-tips/ | the item-size limit is 1 TB; the same page recommends 1000 files or 500GB, and two other help pages state the single-file ceiling and the daily file cap |
| 2026-11-01 | help.archive.org/help/accounts-a-basic-guide-2/ | registration and login take an email address and a password; no institution or document is named anywhere in the account flow |
| 2026-11-01 | help.archive.org/help/uploading-a-basic-guide/ | uploading requires a free account; the signup form and the credentials page state the fields and the key pair |
| 2026-11-01 | help.archive.org/help/uploading-tips/ | 500 GB is a recommendation tied to 1,000 files, not a hard cap; the bulk-upload page states a stricter 500 files or 500 GB |
| 2026-11-01 | help.archive.org/help/uploading-tips/ | the limits as served on 2026-11-01 |
| 2026-11-01 | help.archive.org/help/uploading-what-is-not-ok-or-not-ok-to- | the daily file cap the limits note records, with a zip counting as one file |
| 2026-11-01 | archive.org/developers/ias3.html | the live IAS3 page still states world-readable files and fake ACLs, so the egress terms rest on a page that still serves, not only on the archived copy |
| 2026-11-01 | help.archive.org/help/uploading-troubleshooting/ | the 10,000-file recommendation and the 1 TB item ceiling, and on the same page the 250,000-file API allowance and the 500-700 GB single-file ceiling |
| 2026-11-01 | probe: https://archive.org/download/nasa/__ia_thumb.jpg | an anonymous download of a public item file answers 200 after a 302 to a storage node, 5,399 bytes, no credentials |
More fields 1
| verified | [{"url": "https://help.archive.org/help/uploading-tips/","date": "2026-10-22","via": null,"http_status": 200,"bytes": 5433,"sha256": "48c2cd2e54f5effa68cd5fced7c6b5fb3d5045a2323827fd720e78345b324a82","pattern_found": true,"what_changed": "no change: the page still states the claim","capture": null}] |
|---|
Unknown values 3
- per object cap notes: the field may be stated in a source that was not read
- retention notes: the field may be stated in a source that was not read
- retention stated: the field may be stated in a source that was not read