Albums and files

For Bunkr, GoonBox, CyberDrop, Filester, Pixeldrain, and Turbo, every file in an album or folder downloads concurrently into a subfolder named after the album ID:

~/Downloads/
  k4OuX4pJ/        ← album subfolder
    image1.png
    video1.mp4

Single-file links (such as CyberDrop /f/ or Filester /d/) download straight into the output directory. In a batch each URL gets its own subfolder, the same as albums.

Concurrency and politeness

Setting	Default	Effect
--concurrent	3	Downloads running at once
--per-host	2	Downloads hitting one server (CDN node) at once; 0 = no limit
--limit-rate	unlimited	Combined speed across all downloads, e.g. 2M
--scrape-delay	none	Pause between album scrapes, e.g. 500ms
--max-album-items	10000	Cap items from one album; 0 = unlimited

When a server answers 429 or 503, every download from it pauses for the Retry-After time (or the retry backoff) instead of retrying in a burst. If a host such as a Bunkr CDN node still returns many 503s, try --per-host 1.

Resuming and existing files

Interrupted downloads keep validated .part files and resume with HTTP Range requests, including across runs.

Existing output files are reused only when their source and content can be verified. Unverified files are left alone, and the new download is saved under a non-clashing name.

With --checksums, a sha256sum-compatible <file>.sha256 sidecar is written next to each download. Re-running then verifies existing files and re-downloads any that no longer match. Without it (the default), existing files are checked by size only.

Pixeldrain limits

Pixeldrain throttles speed after about 6 GB per IP in 24 hours. Use --proxy to switch IPs; resumed downloads pick up where they left off.