A public, self-hosted look at the Radicle network, and at the repositories this seed stores for it. Compiled only from the seed's own routing table, server logs and storage scans · no trackers, no cookies, no third parties.
- Half the network is younger than a year. Of the repositories on the network today, 70% were created in the last twelve months, 7,970 of them, up from 2,973 the year before. This is a network still filling up, not an archive being tended.
- The whole public network fits on a laptop. Mirroring every public repository on Radicle takes about ~112 GiB. A few dozen full-history mirrors (of linux, HardenedBSD, nixpkgs, godot, guix and more) account for 52% of that, and 76% of repositories are under 1 MiB. Hosting the commons is within reach of one person and one spare disk.
- Every public repository is backed up many times over. The average repo is carried by 10.1 nodes, and only 466 of 14,893 sit on a single one. No machine here is the machine a project depends on, which is the whole point of putting code on a peer-to-peer network.
- This seed carries most of it, which is why these numbers describe the network. It holds 11,445 repositories, about 77% of the 14,893 public repos it can see, across 1,965 announcing nodes. Everything below is measured from that vantage point, and says so wherever it matters.
01Replication & network reach
Radicle is peer-to-peer: every repo can live on many nodes at once, and the more nodes carry it, the more resilient it is. This is how replication looks across the 14,893 known public repositories and 1,965 nodes the network currently spans.
This seed's policy is to mirror the whole network; the ~23% of public repos it does not hold are mostly the stale, abandoned and testing repos auto-pruned by radicle-seed-prune (it would otherwise sit near 99%), plus a few not-yet-fetched or unreachable ones. Most of the 1,965 nodes are small personal seeds; a few dozen larger seeders carry the bulk, and this one is among them. 466 repositories appear on only one node in this routing view; for those, a seed like this can be the difference between a project staying reachable and quietly disappearing.
The wider network this seed sees
The peer-to-peer network this seed can see spans 3,429 nodes, but only 1,965 actively announce repositories and just 45 are large public seeds carrying most of it. It leans heavily private: hundreds of nodes connect over Tor or Tailscale, and of the few exposing a public address, most sit in Germany, the US, and the rest of Europe.
Collaboration: issues & patches
Counts here are per Radicle identity (one signing key per device). Multi-device identities don't exist yet, so one person can appear as several.
Radicle collaboration is code-centric but balanced: issues and patches run roughly neck-and-neck, and only a small fraction of seeded repos carry any native collaboration. Discussion concentrates in a few actively-developed projects led by heartwood; the long tail of mirrors holds code but little in-repo activity. Filtered by identity, most of that volume is automation: 89.3% of all issues and patches come from CI brokers and mirror bots, not people.
Automation is a recent arrival, and it did not displace anyone: it shows up in September 2024 and human volume simply carried on, growing steadily from roughly 107 to 837 issues and patches a month. The share view is the one to read for what changed, and the volume view for what did not.
Collaboration has a committed core and a long tail of drop-ins: of 936 people who have ever opened or touched an issue or patch here (automated accounts excluded; 974 including them), only about 300 were active in the last three months and 128 in the last month. The rest contributed once or twice and moved on, the usual shape of open-source participation.
02The repositories we seed
This seed mirrors public repositories from across the Radicle network so they stay available even when their authors are offline. Every repo links to its page on app.radicle.at.
Half of all repositories are under 0.23 MiB (effectively just metadata and a few commits), while a small number of full-history mirrors of major projects carry a disproportionate share of the bytes.
03Traffic & visitors
The seed answers two kinds of web traffic: the explorer at app.radicle.at (the pages people open) and the API at seed.radicle.at (what the explorer and other tools call). Both are dominated by automated clients.
By volume the public web surface is mostly machines: search engines indexing repos, plus a constant background of automated scanners probing for things that do not exist here. Stripped of that noise, genuine human visits are a steady few thousand a month.
04What gets browsed
Of the thousands of repos seeded, only about 1,370 were ever opened in the explorer. Repositories featured on the landing page draw the most attention.
The featured set, led by editor integrations like radicle-vscode-extension and radicle-jetbrains-plugin, dominates explorer views by design, the most visible placement on this explorer. Switch to Organic to see what visitors discover on their own.
05Who visits
Based only on the standard User-Agent string every browser sends and the request's network origin. No fingerprinting, no accounts, no cross-site tracking. Everything in this section counts human visits only; crawlers and scanners are excluded.
06Speed & reliability
The explorer is a static app and serves instantly. The API is fast for normal use, with a long tail when crawlers walk the full git history of the largest mirrors.
Adding this always-on seed helps your repos and their updates propagate faster across the network. Any public repo you publish gets mirrored here too, an extra always-online copy that others can still fetch when your own node is offline. It also speeds up fetching the repos this seed already holds.
z6MkjDYUKMUeY58Vtr8dGJrHRvnTfjKWVGCBYJDVTHXsXzm5@seed.radicle.at:8776Connect once from your terminal:
rad node connect z6MkjDYUKMUeY58Vtr8dGJrHRvnTfjKWVGCBYJDVTHXsXzm5@seed.radicle.at:8776Or make it permanent by adding it to preferredSeeds in ~/.radicle/config.json:
"preferredSeeds": [
"z6MkjDYUKMUeY58Vtr8dGJrHRvnTfjKWVGCBYJDVTHXsXzm5@seed.radicle.at:8776"
]Support radicle.at
This is one of the oldest and largest public seeds on the Radicle network, online since early 2024 and carrying about 77% of the network's known public repositories. It keeps projects reachable and quietly powers the explorer and API that the ecosystem and third-party tools rely on, around the clock, for free.
Running it costs real disk, bandwidth, and time. If this seed is useful to you or to the network, a small contribution keeps it healthy and growing.
Donate to keep it flourishingThe button goes to Liberapay, which handles cards and recurring gifts.
DE63 1001 1001 2624 5948 04Account holder Konstantinos Maninakis. Inside the euro area an instant transfer arrives the same minute and costs nothing. From outside EU use BIC NTSBDEB1XXX.
bc1pemmvu4325yuqrn8y44f6rllwp4hqp7znrv3qsr8yv5vx8la3etvqylwnleEvery bit helps cover disk, bandwidth, and upkeep. Thank you for keeping the Radicle network resilient.
- Everything here is aggregated from the seed's own web server access logs and periodic scans of its Radicle storage. Nothing else.
- Zero tracking. No analytics scripts, no cookies, no pixels, no third-party services, no fingerprinting. The only data is what any web server records to answer a request.
- Logs are processed under GDPR: visitor IP addresses are used only transiently to group requests and map approximate country during aggregation, and are never published here. Only counts, sizes, and coarse breakdowns are shown.
How each number is worked out, and where it could be incorrect
- What is missing. Repos fetched or synced directly between nodes, including over Tor or other private transports, leave no trace here. "Known to the network" (14,893) exceeds "seeded here" (11,445) because the routing table tracks repos this node does not fully store.
- People vs. robots is inferred from request patterns and User-Agents. Sophisticated scanners that imitate browsers mean human counts are, if anything, slightly overstated.
- Human vs. automated collaboration is inferred per identity from activity patterns: identities with hundreds of issues or patches arriving seconds apart, or that only ever file patches in bulk, are classed as automation (CI brokers, mirror bots), refined by a hand-checked override list for borderline cases. An issue or patch counts as automated only when every participant on it is automated.
- The monthly people-vs-automation series places each issue or patch in the month it was last active, because the seed holds only the current tip of each thread and not its full edit history. A long discussion counts once, in the month it went quiet, so recent months are the most complete and earlier ones understate whatever is still being worked on today.
- Repos created per month dates each repo by the commit at the root of its Radicle identity, which is the creator's own clock and occasionally absurd, so dates before 2023 or in the future are dropped. It counts only repos this seed still stores, so it is a lower bound: the ones it never fetched, or has since pruned, are missing, and those skew old, which understates the early months the most. It is built by dating this seed's current shelf backwards, not from any record of what the network held at the time, so the running total can only ever rise. Nothing is ever deleted from Radicle, but a repo stops being counted here once this seed no longer carries it. Read it as "when were the repos sitting on this seed today created", not "how big was the network back then".
- The "Freshly created repos" list is the newest edge of that same data, with the same caveats, and it lists what reached this seed: a repo created on the network but never fetched here does not appear, and one that arrives late appears at its creation date rather than at the top.
- Replication and node counts come from this node's routing table at snapshot time and shift constantly. Nodes with restrictive seeding policies announce fewer repos than they hold, so "largest seeder" rankings are a lower bound.
- "Official Radicle" repos are a hand-curated list of genuine Radicle-team projects (core protocol, apps, and the CI ecosystem), maintained by editorial judgment rather than inferred from signing-key identity, since a maintainer's personal repos share the same delegate.
- The headline ~112 GiB "to mirror every one of them" is an estimate, not a measurement: the exact size on disk of the repos this seed does hold, plus the ones it does not priced at the median repo size (0.23 MiB). It should land close, because a few dozen large repos this seed already carries dominate the total (the top 1% hold 52% of the bytes) while the repos it lacks are overwhelmingly the tiny stale and testing tail, but a single unseeded whale would move it.
- Visitors and traffic. Country is approximate, from an offline IP database, by distinct IP; one network address is not one person. Repository view counts reflect full page loads only. Response times are measured at the seed itself, so requests answered from a browser or upstream cache never reach it and are neither counted nor timed.
- How far back it goes. The node has run since 15 March 2024. Repository stats begin 16 March 2024, web logs April 2024, and the storage-size series late May 2024.
radicle.at · an independent Radicle seed · last updated · built with open data from the node itself.
seed.radicle.at and stats.radicle.at are independent community projects and not affiliated with the Better Internet Foundation or the Radicle Team.