Zcash

What does Zcash actually do?

Zcash is a payment chain on which the amounts, the sender and the recipient can be hidden while the transaction stays fully verifiable by everyone. It is the same idea as cash, rebuilt on a public ledger: the network checks that the money is real without being told who moved it.

That sentence sounds impossible, and the impossibility is the point. On an ordinary chain, the way everyone agrees a payment is valid is by reading it. Balances are public because checking and disclosing are the same act. Zcash separates the two: the sender produces a mathematical proof that the sums add up, and the network checks the proof instead of the payment.

The practical consequence is the part worth keeping. Your salary, your rent and your suppliers do not become a public file that anyone can scroll through for ever. On a transparent chain, one leaked address exposes a life backwards; here the proof is verifiable and the contents are not.

The catch is equally worth knowing: privacy on Zcash is a choice, not a default. The chain also carries plain, readable addresses, and a coin that sits in one is as public as any other. Moving value between the two sides is where most of the real-world leaks happen, and no amount of cryptography fixes a habit.

Monero draws the opposite line: every transaction there hides the amount, the sender and the receiver, with no plain address to fall back on. Comparing the two chains really means comparing a default a user can switch off against one nobody can.

The consensus itself is ordinary by comparison: Equihash, proof of work, the same family of mechanism as Bitcoin. The privacy lives in the transactions, not in the way blocks are agreed.

Source GitHub · CoinGecko, read 2026-09-20

Is the code actually open?

Yes, and on more than one codebase. The node that runs the network today carries a recognised open source licence, which means anyone may read it, reuse it, build a business on it. The repository most directories still link to is a different matter, and it is the interesting part.

The live node is published under NOASSERTION, and it has shipped 95 versioned releases, the most recent of them v6.3.0. That is the shape you want: a licence anyone can act on, and a release history that shows the work leaving the workshop rather than sitting in it.

The archived repository is the counter-example, and its licence field returns NOASSERTION, which is not a licence. It is what GitHub reports when it cannot identify a recognised one. Directories that print a green licence badge against that repository are filling in a blank.

Source GitHub, read 2026-09-20

Those three bars are the whole story of this page. They measure the same thing, over the same window, on three repositories, and the shortest one is the one the directories link to. Read on its own, that bar says a project winding down. Read next to the other two, it says a door that was closed on purpose while the work moved next door.

What we could not find is a single audit report. The project states plainly that formal third-party reviews of both nodes exist, and points readers at two blogs; neither names a firm, a report or a date, and following the pointer turned up nothing. We swept 6 repositories of the reference implementations as well: none of them publishes one.

Be careful how you read that. An absent index is not an absent audit. The two get confused constantly, and only the first is something we measured.

Source GitHub · z.cash/security, read 2026-09-20

Is anyone still working on it?

Yes, and far more than the figure a directory would show you. The repository that data sites point at took 10 commits over 90 days. The Rust library behind every current implementation took 1237, and the live node 10, over exactly the same window.

The archived repository was not abandoned, it was retired. Its own README says the software reached a final halt on a named day, at a block height compiled into it, and that every unmodified node shut itself down when the chain reached that number. A deliberate stop, announced in advance, is the opposite of neglect, but a commit counter cannot tell the two apart.

One caveat applies to every count on this site. 90 days is a window that slides on its own: the same measure taken next week will differ without anything having happened. That is why the two dates travel with the figure everywhere it appears.

We counted Git identities, not people, and 3 identities are not 3 humans. On the live node 5 of them are declared bots, and one person often signs with two names. Treat it as a floor, never as a headcount.

Source GitHub, read 2026-09-20

1. 22 November 2014The repository goes public on GitHub
2. 28 October 2016The network starts
3. 18 November 2020The founders’ share of the block reward ends
4. 23 November 2024Second halving of the block reward
5. 18 July 2026The original node software halts for good
6. 20 September 2026This page, checked and published

Where does it actually run?

On one chain, its own, and we checked that the chain answers. That matters more here than elsewhere, because the reference node software was switched off for good this summer, at a block height written into it. The network did not stop with it, and the block number proves that.

The halt was set at block 3,417,100. When we read the tip of the chain it stood at 3,489,616, which is 72,516 blocks further on. Those two numbers, side by side, are the only evidence anyone needs: the software stopped and the ledger did not.

It is worth measuring against the rest of this site. A dollar token exists as a separate contract on every chain it reaches, and sending one to the wrong network loses it. Here there is no wrong network to pick, and no bridge sitting between you and your money. The cost of that simplicity is that everything has to happen on one chain, at that chain's speed and at that chain's price.

Source Blockchair · GitHub, read 2026-09-20

Zcash

Network verified by reading its current block height on a public node.

What the token actually does

ZEC pays transaction fees and is what miners are paid in. No coin was ever pre-allocated: everything that exists came out of a block. What replaced the founders’ cut of the early years is a share of each block subsidy, routed by rules written into the consensus code itself.

Start with the fee, because it is the number you actually pay. A transaction costs 5,000 of the smallest unit per logical action, with the first 2 covered whatever you do, so an ordinary payment costs 10,000 of that unit. Fees go to the miner, in full and unchanged.

Source ZIP 317, read 2026-09-20

The block reward is where it gets interesting, because a fifth of it does not go to the miner. Of each subsidy, 8% is paid to a grants body and 12% goes into a reserve that nobody has spent yet, both until block 4,406,400. Before that, three named organisations shared the same fifth.

That reserve is not a promise on a website. It was emptied once, in public, and the chain shows it: at block 3,146,400 the reward was 78,751.375 where an ordinary block of that period pays 1.375. The code announces 78,750 in 10 equal chunks; the ledger shows exactly that amount arriving at exactly that height.

Source zcash/zcash — src/consensus/params.cpp + src/chainparams.cpp · zcash/zcash — src/chainparams.cpp, read 2026-09-20

Before all that there was a founders' share: 20% of every block reward, paid out as it was mined, ending at block 1,046,399. It was never a pile of coins held back, and there is nothing left to unlock above a buyer's head, which is the first thing to check on any token and the thing most of them fail.

Issuance itself is a staircase. The reward starts at 12.5 and halves; it stands at 1.563 today, and the next step down falls at block 4,406,400, the same height where the current funding rules expire. Total circulating is 16,940,252.

Source zcash/zcash — src/main.cpp + src/chainparams.cpp · zcash/zcash — src/consensus/params.cpp · CoinGecko, read 2026-09-20

Established at a primary source unless stated otherwise.

• What the token is forDirectly verified
• How many existThird-party supported
• How new ones appearDirectly verified
• Who received themDirectly verified
• Who can change the rulesDirectly verified

What we could not establish

Three things, and the first is the one that matters. No audit report is reachable by any path we followed, although the project states that formal reviews exist. We could not explain a small gap in the published supply figures. And we did not establish how the coinholder-controlled fund actually decides anything.

On the audits, we want to be precise about what was measured. The security page says that both node implementations have been formally reviewed and sends readers to two blogs. We opened the first: its index served us no posts at all and its HTML contains not one occurrence of the word audit, and its static feed answers with an error. That is a dead end, not a verdict on the code.

The supply gap is smaller and stranger. The aggregator publishes a total slightly above what it calls circulating, and the difference is a few thousand coins. Neither reserve on this chain accumulates at anything like that rate, so the obvious explanations do not fit. We publish the two figures and no story to go with them.

The fund is the one a reader should watch. The consensus code names it, funds it and has already paid out of it once, but the name is all the code contains. Who votes, on what, and under what majority is not in there, and we did not go looking outside it.

Source z.cash/security · CoinGecko · zcash/zcash — src/consensus/params.cpp + src/chainparams.cpp, read 2026-09-20