DOG Mode Explained: The Bitcoin Client Bypassing BIP-110 Constraints

BIP-110 aims to limit non-financial data on Bitcoin through a consensus change, but it has drawn almost no miner support. A proposed DOG Mode client, however, takes the opposite approach—loosening restrictions without requiring any vote.

The idea gained attention shortly after reports showed that efforts to remove non-financial data were nearing their deadline with virtually zero backing from miners. Instead of pushing for consensus, some developers are now exploring a path that sidesteps it entirely.

Leonidas, co-founder of Runestone and a prominent figure in the Ordinals and Runes ecosystem, has put forward the concept of an open-source Bitcoin client called DOG Mode.

The client would remove two key limits enforced by Bitcoin Core, the software run by most nodes. One caps the size of transactions that nodes will relay, while the other sets a minimum value threshold for outputs.

Consensus rules determine whether a block is valid, and violating them would split a node from the network. Relay policy, by contrast, governs which transactions nodes choose to forward. Bitcoin Core filters out “non-standard” transactions, even if they are valid under consensus rules.

Because most nodes run Core, these defaults effectively shape Bitcoin’s behavior. Miners can still include such transactions if they receive them directly, but Core does not relay them, forcing users to submit them straight to miners. Services like MARA’s Slipstream exist to handle these larger or more complex transactions.

DOG Mode would raise the maximum standard transaction size from 400,000 weight units to 3.9 million. Since a Bitcoin block can hold up to four million weight units, Core currently relays transactions that fill only about 10% of a block, while DOG Mode would allow transactions that nearly occupy an entire block.

It would also reduce the dust limit—the minimum output value worth relaying—from roughly 294–546 satoshis to just one satoshi, the smallest unit of Bitcoin.

Protocols like Ordinals, which embed images and text into transactions, and Runes, which enable token issuance on Bitcoin, currently require extra bitcoin to meet this threshold. Leonidas argues that removing the dust limit could free up roughly $25 million locked as padding in these systems.

BIP-110, a user-activated soft fork, requires 55% miner approval. Yet it has failed to gain traction, recording no support in the current period and never exceeding about 1% historically. This underscores the lack of momentum behind the proposal.

DOG Mode, by contrast, does not rely on network-wide agreement. It simply changes what individual nodes relay. If enough nodes adopt it and at least one miner accepts the fees, those transactions can still be confirmed—without any signaling requirements or deadlines.

Support for BIP-110 remains minimal and is largely concentrated among users of Bitcoin Knots, an alternative client favored by those pushing for stricter data limits. DOG Mode would represent the opposite approach—a Core-based fork designed to relax restrictions while remaining closer to the original software than Knots.

For now, DOG Mode exists only as a concept. There is no codebase, repository, or official release. Leonidas has called on developers to build it, miners to support it, and users to promote it, leaving the project in its earliest stages.

Leonidas, who helped popularize the Runes protocol and co-founded Runestone—the project behind the DOG token—is now proposing this client while relying on others to develop it. Notably, the proposed changes could release funds back into markets where his token is actively traded.

In short, BIP-110 seeks to rewrite Bitcoin’s rules and requires broad approval it does not have. DOG Mode, meanwhile, only changes node behavior and would need just one miner willing to process the transactions. One depends on consensus; the other bypasses it entirely.

DOG token prices were largely unchanged following the announcement, slipping about 1.2% over the past 24 hours.