Search Marginalia's independent web index, which prioritizes noncommercial pages and includes tools for finding older sites.
Marginalia Search
Marginalia Search indexes text-oriented websites, including older and less prominent pages from across the web.
About Marginalia Search
Marginalia Search is an internet search engine for text-oriented websites. It indexes the small, old and less prominent parts of the web, with the stated aim of developing alternative discovery methods and elevating non-commercial parts of the Internet. The project describes itself as both an experimental workshop and a public service.
The source code can be run as a copy of Marginalia Search or as a white-label search engine for the operator's own data, which may be crawled or side-loaded. Self-hosting requires a setup step in which a script downloads model data before compiling. The README describes a barebones self-hostable mode and links to further documentation. A production-like deployment needs substantial RAM, fast SSDs and several terabytes of hard drive storage. The README says it can run on smaller hardware by limiting index size, and that it will run on a 32 GB machine, though performance may be poor at that size. The search logic is not very configurable at present, and many ranking judgements reflect the project's own goals.
Key features
- Indexes text-oriented websites, including older and less prominent pages
- Can run as a copy of Marginalia Search or as a white-label engine
- Supports white-label use with crawled or side-loaded data
- Self-hostable barebones mode with a setup and operation demo
- Setup script downloads model data needed before compiling
- Source code published on GitHub with contributor documentation
Best for: Developers and organizations that want to self-host a text-focused search engine, or run one on their own crawled or side-loaded data.
Good for
- Designed to run on affordable hardware with minimal operational overhead
- Documented setup path, including a demo of self-hosted operation
- Funded in part through an NLnet NGI0 Entrust Fund grant
Watch out for
- A production-like environment requires a lot of RAM and large storage
- The README says performance may be poor on a 32 GB machine
- Search logic is not very configurable at present
Marginalia Search at a glance
- Category
- Search & Reference › Search Engines
- Type
- Open-source project
- Pricing
- Not confirmed. Check the site
- Open source
- Not stated
- Language
- English
- Last checked
- Oct 6, 2026 · Online when checked
- Listed
- Oct 5, 2026
Marginalia Search alternatives
Tools and sites you could use instead, ranked by how closely they match.
Wiby searches for older-style web pages by subject and lets visitors submit pages for its index.
Popular in Search Engines
Well-known listings from the same topic.
Bingbing.comLilo,lilo.orgSearch My Sitesearchmysite.netSearch craigslistsearchcraigslist.orgswisscowsswisscows.comMetaGermetager.org
Marginalia Search FAQ
Can I run my own copy of the search engine?
Yes. The system can be run as a copy of Marginalia Search, or as a white-label search engine for your own data, which can be crawled or side-loaded.
What hardware is needed?
A production-like environment needs a lot of RAM, ideally enterprise SSDs for the index, and additional terabytes of slower hard drives for crawl data. It can run on smaller hardware by limiting index size, and a local developer deployment is possible with much smaller hardware.
Can the search behavior be configured?
Not very much at present. Many of the judgements are based on the Marginalia project's goals, and additional configurability is stated as being worked on.
Is something wrong or out of date?
Tell us and an editor will check it.
Own this website?
Claim this listing
Claim the listing and suggest updates yourself. An editor checks every claim first.