Bitcoin Knots

Knots is not a competing currency or a rival chain. It began in 2011 as a patch set called Bitcoin LJR, it is rebuilt from each Bitcoin Core release, and its version strings say so: v29.4.knots20260508 is Core 29.4 plus Knots patches dated May 8, 2026. What it changes is policy, the set of transactions a node is willing to pass along to its neighbors.

How it works

Knots takes Bitcoin Core's source, applies patches, and ships the result with different opinions preinstalled.

The validation code underneath is the same. Knots computes the same block hashes, enforces the same signature rules, applies the same subsidy schedule and accepts the same chain. Nothing in it can create a coin Bitcoin Core would reject or reject a block Bitcoin Core would accept, because none of the patches touch consensus.

The differences are upstream of the chain. Knots has long capped OP_RETURN data at 42 bytes by default, against the 83 that Bitcoin Core carried for a decade and the 100,000 that Core adopted in version 30.0 on October 13, 2025. It adds options for refusing transactions its operator considers abuse of the chain rather than payment, and it exposes more configuration than Core does, on the principle that the operator should be able to decide.

Understanding what that achieves is the important part. A relay rule is a filter on your own inbox and outbox. If a miner includes a transaction your node declined to forward, your node validates the block, accepts it, and stores that transaction forever. Filtering shapes what travels the public network and makes unwanted data slightly more expensive to place. It does not remove anything from the ledger, and enough operators filtering the same thing pushes the traffic toward direct arrangements with mining pools rather than out of existence.

The project's honest weaknesses are structural. It has one principal maintainer and a fraction of Bitcoin Core's review capacity, and every patch is code that Core's contributors have not signed off on. Dashjr is a veteran contributor and a BIP editor, which is why the project is taken seriously, but a small team is a small team.

Where you see it

Knots turned from a specialist's client into a visible bloc during the argument over data storage on the bitcoin chain.

Public node counts put it on about 12 percent of reachable nodes in June 2025, 19 percent that September, and roughly 22 percent by late July 2026. Those figures count reachable nodes only, and node counts are easy to inflate, so treat them as a direction of travel rather than a census. The people switching were largely objecting to inscriptions and to Bitcoin Core's decision to stop restricting OP_RETURN.

For someone whose interest in bitcoin is buying and holding it, nothing here demands action. Both clients verify your coins identically. Choose Knots if you want your machine to refuse to carry data you regard as spam and you are comfortable with a smaller review process. Choose Core if you want the larger one. Choose neither and you are trusting a company's server instead, which is the only option on the list that actually weakens your position.

Bitcoin Knots vs a hard fork

Switching to Bitcoin Knots is not forking bitcoin, and the distinction is the one most often mangled in coverage of the dispute. A hard fork changes the rules that decide whether a block is valid, so nodes running the old software reject the new chain and two coins exist, which is what happened when Bitcoin Cash split away in August 2017. Knots changes nothing about validity. Its users and Bitcoin Core's users sit on the same chain, see the same balances and can pay each other all day. They disagree only about which pending transactions they are prepared to hand to a peer, which is a preference, not a schism.