What Is an NZB File? How NZB Files Work on Usenet

An NZB file is a small XML document that acts as a map to content on Usenet. It contains a list of article Message-IDs, the newsgroups those articles were posted to, and the segment numbers that make up the original file. When you hand an NZB file to your download client (SABnzbd or NZBGet), the client uses those Message-IDs to request each segment from your Usenet provider, then reassembles the original file.

Think of it like a table of contents. The NZB file does not contain the actual content. It tells your download client exactly which articles to fetch and in what order. A typical NZB file is a few kilobytes in size, even if the content it points to is many gigabytes.

An NZB file is plain XML. If you open one in a text editor, you will see something like this (simplified):

<nzb>
  <file subject="example.rar (1/50)">
    <groups><group>alt.binaries.test</group></groups>
    <segments>
      <segment number="1">[email protected]</segment>
      <segment number="2">[email protected]</segment>
    </segments>
  </file>
</nzb>

Each <segment> contains a Message-ID. Your download client sends these Message-IDs to NewsDemon, and the server returns the corresponding article data. The client collects all segments, decodes them from yEnc encoding back into binary data, and reconstructs the original file.

With SABnzbd

Drag the NZB file onto the SABnzbd web interface, or click the "Add NZB" button and browse to it. SABnzbd immediately starts downloading the segments listed in the file. When all segments arrive, SABnzbd runs PAR2 verification, repairs if needed, and extracts the final files.

With NZBGet

Same process. Upload the NZB through the web interface or drop it into NZBGet watched folder. NZBGet handles the rest.

With automation tools

If you use Sonarr, Radarr, or Prowlarr, you never handle NZB files directly. The automation tools search indexers, select the best NZB, and send it to your download client via API. The NZB is still the mechanism underneath, but the process is invisible to you.

NZB files are created by NZB indexing services. These services scan Usenet newsgroups, catalog the articles they find, and generate NZB files that point to specific content. When you search an indexer and click download, you get an NZB file.

NZB indexers are separate from Usenet providers. Your provider (NewsDemon) stores and serves the actual articles. Indexers catalog what is available and give you the NZB files to find it. You need both: a provider for access and an indexer for discovery.

Some indexers are free with limited features, others require a paid membership for full access. The Usenet community on Reddit and other forums discusses which indexers are reliable.

An NZB file serves a similar purpose to a .torrent file: it tells your client where to find the corresponding articles. The key differences are in how the download actually works.

A .torrent file tells your client to download pieces from other users (peers) over a peer-to-peer network. Your IP address is visible to every other peer. Download speed depends on how many people are sharing the file.

An NZB file tells your client to download segments from a Usenet server over a direct, encrypted connection. No other users are involved. Download speed depends only on your internet connection and your provider. Our comparison page covers the full set of differences.

Can I create my own NZB files?

Yes. When you post files to Usenet, your posting tool (Nyuu, NewsUP) can generate an NZB file after posting. This NZB contains the Message-IDs of every article you posted, which you can share with anyone who wants to download the content.

Do NZB files expire?

The NZB file itself does not expire, but the articles it points to might. If the articles are older than your provider retention window, they may no longer be available. NewsDemon offers 5,695+ days of retention, so NZBs pointing to content posted within the last 15+ years should still work.

Are NZB files legal?

An NZB file is just a list of Message-IDs. It contains no copyrighted content itself. The legality depends on what the articles it points to contain, the same way a URL is not illegal but the page it links to might be.