the delivery layeranyone can servepriced, not quoted.
the first bytes of a 14-gigabyte file posted in berlin reach a client in tokyo in under a second. the client streams from three peers at once, verifies every chunk with blake3, and pays per megabyte in usdc — whether the payload is a linux iso, a dataset, a game patch, a media library, or an ai model. deCDN is demand-shaped, locality-optimised delivery for large files at scale: supply forms around demand, cost collapses as regional traffic concentrates. the code is open. the network is open. the price is posted.
information scaled.supply didn't.
the pattern repeats whenever something big ships: mirrors fork, cdns rate-limit, small teams burn tens of thousands hosting bytes they don't own. deCDN inverts every axis — supply forms around demand, not allocated to it.
How it works
in a single handshake, the client asks nearby peers who has the file. peers answer with what they've cached, their rate per megabyte, and how fast they can serve — the roundtrip averages under 100 milliseconds. the client ranks the answers by price, latency, and reputation; the best-priced, fastest, most-reputable peer wins, or several win in parallel for a large file.
bytes flow directly from the chosen node; for files over ten gigabytes the client opens parallel streams to several peers at once and aggregates their throughput — a 1 gbps origin turns into multi-gigabit delivery to the client. every chunk is verified against the blake3 tree hash the instant it lands; tampered bytes trigger immediate disconnect and a fraud proof against the node's stake. trust no node — verify every byte.
you pay per megabyte in usdc, automatically, as the bytes arrive — no monthly invoice, no subscription, no whole-file minimum. pay for what you pulled, nothing more.
frequently asked.
It's real, and for a small project it's the right call. But that free egress is a revocable vendor arrangement, not a guarantee: R2 is Cloudflare's own loss-leader, repriceable at its discretion, and B2's free egress rides the Bandwidth Alliance, a pact among rivals that any member can exit. Either way, you depend on one provider's roadmap, terms, and outage surface. deCDN is the opposite: an open market where operators set their own rates and compete for every request, so the price falls as the network grows, on capacity that's already paid for at the edge rather than in hyperscale datacenters. We don't claim to beat free on the egress line, and no token subsidy props up demand. Against a CloudFront-plus-S3 bill today ($0.04 to $0.20 per gigabyte, list pricing), deCDN is materially cheaper: around $0.01/GB and trending down.
Every blob is BLAKE3-addressed, so the hash the client is asking for is the same hash they'll be checking on the way in. Verification happens chunk-by-chunk, in flight — mismatched bytes are dropped and the next peer on the list is asked instead. Peers that send garbage don't get paid for it: the mismatch is caught in flight and the bytes dropped. On-chain misbehavior (phantom availability, rate manipulation, blacklist violations) is slashed against the peer's capacity bond at submit time, from an EIP-712 signed message rather than a central arbiter. The math is the arbiter.
Both. For gating: upload ciphertext instead of plaintext — nodes cache the encrypted blob without ever seeing inside, and your app holds the keys, handing them to clients over a separate authenticated channel. Subscription, paywall, whatever logic fits. For takedown: governance can flag specific BLAKE3 hashes as non-servable, and nodes that keep serving them past a short compliance window lose part of their capacity bond — slashing is on-chain, not a human call. Both the key gate and the flag list are auditable; compliance runs end-to-end, from origin onboarding to peer-level slashing, governance-owned and auditable throughout.
TOKEN secures the network — operators post a slashable capacity bond to participate, and they lose it if they misbehave. Payments, though, happen in USDC, because operators want stable currency they can pay power bills with. Thirty percent of protocol fees flow into a Balancer V3 80/20 TOKEN/USDC buyback-and-burn, tying token demand to network throughput.
Yes. Account abstraction lets a publisher fund a payment channel for their audience, so users pull content with no wallet and no subscription — same free-to-download experience as a public mirror, except the publisher pays peers per megabyte instead of one cloud's egress bill. Default flow is client-pays; publisher-pays is the flag you flip when you want to ship widely without per-user payment friction.
Contact
the network is open. so is our inbox. write us with questions, partnerships, or anything you'd run on a fleet of idle machines. we read everything.