KLEAR3
English

Bitcoin

KLEAR3 ResearchPublished 2026-08-25Updated

BitcoinBitcoinBTC

Bitcoin, the first ledger that works without a company issuing it, has published its code since December 2010. Its supply is not capped at 21 million: the maximum ever programmed is 20,999,999.977, because each halving rounds down to whole satoshis and the remainder is never created.

What does Bitcoin actually do?

Bitcoin is a shared ledger that nobody owns. Anyone can add a payment to it, and no company can refuse one, freeze an account or change the rules alone. What makes that possible is that thousands of machines keep the same copy and agree on it by spending electricity.

The word ledger matters more than the word coin. Bitcoin does not store objects in your wallet; it stores a list of movements that everyone can recompute from the first one. A balance is not a number written somewhere, it is what the list adds up to when you follow it to the end.

The electricity is the part people find absurd, and it is the part that does the work. To add a block you must find a number whose fingerprint starts with enough zeros, using SHA-256. There is no shortcut, only trying. Rewriting old history would mean redoing every trial since, faster than everyone else combined.

So the cost is not a side effect to be optimised away. It is the lock. Whoever wants to change the past has to buy more electricity than the rest of the world, and keep buying it.

That design choice is what separates Bitcoin from every other entry in this top ten. Ethereum stopped spending electricity in 2022 and secures itself with deposits instead. Neither is a lesser version of the other; they pay for the same guarantee with a different currency.

Source CoinGecko https://api.coingecko.com/api/v3/coins/bitcoin, read 2026-08-25

Is the code actually open?

Yes, and unusually so. The code has been public since December 2010 under the MIT licence, one of the most permissive there is: anyone may read it, reuse it, sell a product built on it. Sixty eight versions have been published. Nothing here requires anyone to be trusted.

The distinction that catches most readers out is between published code and open source code. Plenty of projects publish their contracts without granting anyone the right to reuse them. Bitcoin grants it, in writing, and has done since the beginning.

It is worth measuring against the other end of this top ten. Hyperliquid publishes no protocol source at all: its node repository holds a Dockerfile and a signing key. Between the two sits Tether, whose contract has no repository while the company that issues it maintains a hundred and fifty others.

None of that says Bitcoin is safe. An open licence means outsiders can look, not that anyone has. Which brings us to the uncomfortable part of this page.

Source GitHub https://api.github.com/repos/bitcoin/bitcoin, read 2026-08-25

How much has already been created
20075100/ 20999999.9769of everything that will ever exist

No audit firm has reviewed this code, and none is listed anywhere. That is not an oversight. Bitcoin never had a company to commission one, and its review model is the opposite: every change goes through a public proposal, argued in the open, for as long as it takes.

What protects the code here is not a report with a date on it. It is fifteen years of running with money on it while anyone in the world was free to look for a flaw and take the proceeds. That is a different kind of evidence, and our rating counts it as zero, because it is not a rating of safety.

Source GitHub, read 2026-08-25

Is anyone still working on it?

More than on any other project in this top ten except one. The repository took 1 192 commits from 19 distinct Git identities over the 90 days to 25 August 2026. The last one landed the day we read it. A project this old is usually assumed to be finished.

The number is worth pausing on because the intuition runs the other way. Bitcoin is the oldest entry here, and it is four times busier than Ethereum's main client, which took 255 commits over the same window. Age and activity are not the same axis.

One caveat that applies to every count on this site. Ninety 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 also counted Git identities, not people. One person often signs with two, and a company can hide behind one. Nineteen identities is a floor on the number of humans, not a headcount.

Source GitHub https://api.github.com/repos/bitcoin/bitcoin/commits?since=2026-05-27T00:00:00Z&per_page=1, read 2026-08-25

  1. The Bitcoin network starts
  2. The repository goes public on GitHub

Where does it actually run?

On one network, its own, and we checked that it answers. A public node returned block 964 047 on 25 August 2026. Bitcoin is the only entry in this top ten that lives on a single chain and has never been issued anywhere else.

That sounds unremarkable until you compare it with the rest of the list. Tether exists as eleven separate contracts on eleven chains, and USDC as thirty four. Sending either one to the wrong network loses it. With Bitcoin there is no wrong network to pick.

Wrapped versions of it do circulate elsewhere, issued by third parties. Those are claims on Bitcoin held by somebody, not Bitcoin, and this page does not cover them.

Source mempool.space https://mempool.space/api/blocks/tip/height, read 2026-08-25

BitcoinBitcoin

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

What the coin actually does

It pays for block space, and it pays the miner who adds the block. There is no company treasury, no team allocation and no unlock schedule, because nobody issued it. New coins appear only as a mining reward, on a schedule written in the code.

The schedule is the whole economics, and it fits in three lines of code. A block pays 50 coins, that reward halves every 210,000 blocks, and the halving is done by shifting bits, not by dividing. Integers have no fractions, so every halving quietly discards the odd satoshi.

Add up the whole series and you get 20,999,999.977 coins, which is 0.0231 short of the round number everyone repeats. The reward reaches zero at the 33rd halving; the code itself guards at 64, because past that the shift would be undefined.

Fees are the other half. Whoever sends a transaction bids in satoshis per virtual byte, and the miner keeps the bid. As the reward shrinks toward nothing, fees are what will pay for security. Whether they will be enough is the open question of the next century, and nobody has measured it because it has not happened.

Established at a primary source unless stated otherwise.

  • What the token is forDirectly verified
  • How many existDirectly verified
  • How new ones appearDirectly verified
  • Who received themDirectly verified
  • Who can change the rulesDirectly verified

KLEAR3 rating

4.6out of 5
Weakest axisAudits3 / 5
  • Open code5

    public repository · MIT licence · 68 published releases · 1192 commits in 90 days

  • Audits3

    code reviewed in the open: 1192 commits in 90 days, 15 years in production

    What holds it back no report in our registry · no report opened: we do not know what they found

  • Business model5

    business model published by the protocol · paid by celui qui envoie la transaction · collected by le mineur qui produit le bloc · intégralement au mineur, rien n’est détruit

  • Governance5

    decided by bip-process, with no vote · 210 improvement proposals published · process open for 12 years

  • Token alignment5

    supply capped at 21 million · 96% already in circulation · emission schedule published · no initial allocation: every coin comes from the block reward

  • Maturity5

    15 years of public code · 1 network verified by us

We rate the protocol: how it is built, how much of it can be checked, how its token lines up. Never an investment. A high score is not advice to buy, and a low one is not advice to sell.

What we could not establish

Three things, and they are the ones a reader should weigh. We did not open an audit report, because none exists. We did not count holders. And the circulating figure is a third-party number we did not verify on the chain ourselves.

The circulating supply of 20,075,100 comes from an aggregator. It is also a ceiling rather than an inventory: coins whose keys are lost are still counted, and nobody can tell which. Estimates of how many are gone vary by millions, so we publish none.

Our rating gives Bitcoin nothing on the audit axis. Read that as what it is, an absence of reports, not a finding against the code.

Frequently asked questions

Is there really a limit of 21 million bitcoins?

No. The number written in the code is not a total at all, and the total it produces is the figure this page opens with. There is no line anywhere that says 21 million; that figure is a rounding of a sum nobody wrote down.

What the code contains is a reward of 50 coins per block that halves by shifting bits every 210 000 blocks. Bit shifting on whole numbers cannot produce a fraction, so each halving drops the odd satoshi for good.

Add the 33 halvings that produce anything and the series stops 0.023 coins under the round number. The difference is worthless in money and revealing in method: the most repeated figure in the industry is an approximation of something the protocol never states, and nobody ever checked.

If nobody audited the code, why would it be safe?

It may not be. What can be said is narrower and more useful: the code has run for over fifteen years with large sums exposed, published under a licence that lets anyone look, and no audit firm has ever been paid to review it.

An audit is a photograph with a date. It says what a team found in the code as it stood that week, and says nothing about every line written since. Several projects in this top ten hold reports that stopped years ago.

What Bitcoin has instead is time under pressure: fifteen years during which finding a flaw paid better than reporting it. That is evidence of a kind, and it is not the same thing as a guarantee. Our rating scores it zero because it measures reports, not survival.

Why does mining have to waste so much electricity?

Because the electricity is what makes rewriting the past expensive. Remove the cost and anyone could rebuild the history cheaply. The energy is not a side effect of the design; it is the thing being bought.

Every block is found by trying numbers until one produces a fingerprint starting with enough zeros. There is no clever route, only volume, which is why the work cannot be faked or shortened.

Undoing an old block means redoing all that work, and outpacing everyone still adding to the chain. The deeper a payment sits, the more it would cost to erase. Solana buys the same guarantee with deposits instead of electricity, which is cheaper in energy and different in what it puts at risk. Neither approach is free; they simply bill in different currencies.

Read next: our XRP profile shows a chain where a third of the supply is issued but not circulating, and our Figure HELOC profile shows the entry in this top ten that is not a currency at all.

Read this page another way

Ask an assistant to summarise it and check what it claims.