The BIP-110 Bitcoin Fork Explained: What the BLAKE2b Chain Is and Why It Exists

If you only follow Bitcoin through price headlines, you may have missed the most consequential thing to happen to the network in nearly a decade. On August 8, 2026, at block 961,632, Bitcoin split. A group of node operators, developers, and small miners activated a rule set called BIP-110, and when the large mining pools refused to follow it, that group did something most people assumed was impossible: it replaced Bitcoin’s mining algorithm and kept going without them.

The result is a second chain that carries all of Bitcoin’s history, every coin, and every key, but secures its new blocks with a different algorithm called BLAKE2b. Its supporters call it Bitcoin. Most of the industry calls it a fork. This article walks through what happened, why, and what it means for you, told sympathetically from the side that did it, with the risks stated plainly at the end.

First, the one-sentence version

Everything else is detail. But the detail matters, because it explains why thousands of ordinary people were willing to walk away from the most powerful mining industry on earth.

The problem: Bitcoin blocks were filling with things that are not money

Every Bitcoin node stores a full copy of the blockchain. That is the source of the network’s strength: nobody owns the ledger because everybody has it. But it also creates an obligation. Whatever gets written into a block must be downloaded, verified, and stored forever by every full node, including the one someone runs on a small computer in a spare bedroom.

For most of Bitcoin’s life, that data was almost entirely financial. Beginning in 2023, a wave of techniques for embedding images, text, tokens, and arbitrary files in Bitcoin transactions changed that. We covered the mechanics in our guide to Ordinals and Runes. Blocks filled with data that had nothing to do with payments, fees spiked for ordinary users, and the blockchain grew faster than it ever had.

Supporters of BIP-110 saw this as a question of purpose. Bitcoin’s white paper describes “a peer-to-peer electronic cash system.” Node operators volunteer their disk space and bandwidth to verify money, not to host a permanent, unpaid file server for whoever bids the highest fee that day. In their view, unlimited arbitrary data does three things: it makes running a node more expensive, which centralizes the network; it crowds out the small transactions Bitcoin was built for; and it forces every node operator to store content they never agreed to hold.

The software split that came first

Most Bitcoin nodes run Bitcoin Core, the reference software. A long-maintained alternative called Bitcoin Knots, led by early contributor Luke Dashjr, keeps stricter defaults about which transactions a node will relay. For years the difference was academic. Then, in late 2025, Bitcoin Core version 30 removed the long-standing default limit on OP_RETURN data outputs, the main door through which arbitrary data enters blocks.

To the money-only camp, that was Core formally giving up. Knots kept its filters, and something unusual happened: node operators switched in large numbers. By 2026, Knots nodes made up a meaningful share of the reachable network, a grassroots preference for a non-Core implementation that Bitcoin had never seen before. If you want to understand what a node actually does and why this matters, start with our beginner’s guide to running a node.

Why a filter was not enough, and what BIP-110 proposed

A node’s relay filter is only policy. It can decline to pass along a spam transaction, but if a miner includes that transaction in a block anyway, the block is still valid and every node accepts it. To actually keep data out, the restriction has to become a consensus rule, something every valid block must obey.

That was BIP-110. Drafted by Dashjr, it proposed a temporary soft fork that would cap large OP_RETURN outputs, restrict the script formats used for embedding, and prohibit contiguous chunks of arbitrary data above 256 bytes. The limits were designed to expire on their own, buying the network time to develop better long-term answers.

Bitcoin soft forks have traditionally activated through miner signaling, and BIP-110 asked for 55 percent of blocks to signal support. It never came close. Roughly 2.5 percent did. The industrial pools, whose revenue increasingly comes from exactly the high-fee data transactions BIP-110 would restrict, declined to vote for their own pay cut.

The user activation, the stall, and the decision that changed everything

Here the story turns on Bitcoin’s deepest design principle: users, not miners, ultimately define the rules, because a block that users’ nodes reject is worthless no matter who mined it. BIP-110 supporters activated the rules on their own nodes at block 961,632 on August 8, 2026. From that point on, any block containing prohibited data was invalid on their chain.

The SHA-256 mining industry did what its critics feared: it ignored the rules and kept producing data-filled blocks. With almost no hash power behind it, the BIP-110 chain produced only eight blocks, heights 961,632 through 961,639, and stalled. For roughly three weeks it looked like the effort had failed.

What followed is the part supporters are proudest of. Rather than accept that a handful of mining companies could veto what a network’s users wanted, the developers concluded that the miners had become the problem and that the answer was to make their machines irrelevant. Dashjr wrote the code to replace SHA-256d with BLAKE2b, a modern hash algorithm that Bitcoin’s ASIC miners physically cannot compute. Supporters called it, only half-jokingly, “firing the miners.”

On August 30, 2026, a miner known as Silent Wave found block 961,640, the first block in Bitcoin’s history not secured by SHA-256. Within days more than 800 blocks followed. Around September 1 the code was folded into an official Bitcoin Knots release so anyone could run it. In the project’s own words, the fork “continues the existing book and changes how the next pages are proven.”

What actually changed on the BLAKE2b chain

For readers who want the specifics, here are the rules as of early September 2026:

• Proof of work: BLAKE2b permanently replaces SHA-256d. Existing Bitcoin ASICs cannot mine it, but older machines built for the Siacoin network can.
• Block header: Grew from 80 bytes to 164, which is why older light wallets and block explorers do not automatically understand the new chain.
• Block size: A temporary cap of 800,000 weight units (roughly 300 kB), about a quarter of the legacy limit, alongside the BIP-110 data restrictions. Both are scheduled to expire September 1, 2027.
• Supply and history: 21 million coins, the same halving schedule, and the same balances as Bitcoin up to the split. Your keys work on both chains.
• Replay protection: Opt-in only, through a new signature format called SIGHASH_UNIFIED. More on why that matters below.
• Software: Bitcoin Knots with BLAKE2b support. Bitcoin Core and older Knots versions follow the legacy SHA-256 chain.

If the idea of one history branching into two chains is new to you, our older explainer on Bitcoin forks covers how the 2017 splits worked. The BLAKE2b fork is unusual in one respect: it is the first to change proof of work itself rather than just block rules.

Why supporters think this is the most Bitcoin thing that could have happened

It is easy to read the fork as a loss, a split, a schism. Its supporters read it the opposite way. Bitcoin’s founding promise was that no company, government, or cartel could dictate the rules. For years the practical reality drifted from that promise: a few large mining pools produced most blocks, and their commercial incentives increasingly shaped what the network would and would not enforce. BIP-110’s failure to reach 55 percent was not a referendum of users; it was a referendum of a few dozen pool operators.

By changing the algorithm, the community demonstrated that mining power is rented, not owned. Miners secure the chain that users value. If users decide to value a different chain, all that expensive hardware secures nothing. That is not a bug in Bitcoin; it is the entire point of the design, and it had never been tested at this scale before. Whether or not the BLAKE2b chain grows large, it has already proven the principle.

There is also a quieter appeal. In its first weeks the new chain’s total hash rate was measured in tens of terahashes per second, less than a single industrial container on the legacy network. That means a single person with an affordable machine can mine blocks and help secure a network in a way that has not been possible on Bitcoin for more than a decade. We cover exactly what that takes in our home-mining guide for the BLAKE2b chain.

What this means for the coins you already own

This is the practical part, and it deserves care. If you held Bitcoin before block 961,632, you now have the same balance on both chains, controlled by the same seed phrase. That is good news, but it comes with one serious hazard: replay.

Because the two chains share history and, by default, the same transaction format, a transaction you broadcast on one chain can be copied and broadcast on the other, spending your coins there too. The BLAKE2b chain offers an opt-in signature type, SIGHASH_UNIFIED, that prevents this, but your wallet has to support it. Until it does, the safest instruction is the simplest one: do not move pre-fork coins on either chain in a hurry. They are not going anywhere. Wait for your wallet or hardware device to publish clear guidance on splitting coins safely, and follow it exactly.

Hardware wallets need a note too. Signing an ordinary transaction is unchanged, but devices that verify chain identity or proof of work require firmware updates to work with the new chain. Check your manufacturer’s guidance before assuming compatibility, and never enter your seed phrase into any tool that promises to “claim” your fork coins for you. That promise is the oldest fork scam there is.

The honest risks

Supporters make a principled case, and this article has presented it on its own terms. A responsible guide also tells you what the chain does not yet have.

• No market price yet. As of early September 2026, no major exchange listed the coin and the big price trackers had not assigned it a ticker (BIP110 has been proposed). Coins you mine are worth what someone will pay for them.
• Thin security. A chain with a small hash rate can be reorganized by anyone who brings more BLAKE2b hardware than the honest miners have. Treat confirmations on the new chain as far less final than on Bitcoin until that changes.
• A contested name. Most of the industry regards the BLAKE2b chain as a new coin and continues to call the SHA-256 chain Bitcoin. In a wallet, on an exchange, or in a contract, “Bitcoin” almost always still means the legacy chain.
• Fast-moving facts. Everything in this article is dated. Verify current details on the project’s official sites before you act.

The bottom line

For most readers, nothing about your plan changes. Your existing bitcoin is safe on the legacy chain and now also exists on a second chain you can ignore, explore, or eventually claim with care. What has changed is the story. A determined group of users just showed that no mining cartel is permanent and that the people running the nodes, not the people running the machines, get the final say over what Bitcoin is. Whatever the BLAKE2b chain becomes, that lesson is now part of Bitcoin’s history, and we think it is worth celebrating.

We added a full chapter on the fork, including a step-by-step mining walkthrough, to the September 2026 edition of our ebook. If you have already bought it, your download link delivers the updated version.