Class Labs, the group constructing the Monad blockchain, launched Cadence on July 7, 2026, a consensus protocol that lets a number of validators suggest blocks directly as an alternative of handing that energy to a single chief.
The corporate introduced it by way of its official X account, alongside a technical paper on arXiv authored by 9 researchers, together with Kushal Babel and Victor Shoup. In simulations throughout Monad mainnet’s 200 validators, Cadence finalized blocks in a median of 219 milliseconds, a pace Class Labs says matches the most effective case of older single-leader methods whereas permitting block intervals as brief as 100 milliseconds.
What Cadence Modifications for Monad
Most blockchains give one validator management over a whole block: it decides which transactions get in and might reorder them earlier than anybody else sees the consequence. Class Labs’ X announcement factors to that management as the basis of most MEV extraction right this moment. Cadence removes it by letting a number of proposers submit block contents in parallel, then merging the outcomes deterministically so no single proposer units the ultimate order.
The arXiv paper describes Cadence’s quick path as ending in three communication rounds, matching conventional single-leader consensus, with a quicker two-round “speculative finality” borrowed from Class Labs’ present MonadBFT design. Every block runs as an unbiased consensus occasion underneath what the researchers name excessive pipelining, decoupling block interval from community delay.
Class Labs additionally linked Cadence to BTX, its parallel design for encrypted mempools, calling the pairing a step towards fixing MEV contained in the protocol somewhat than patching it afterward.
What This Means for Monad Merchants and Builders
Cadence targets the precise mechanic behind front-running and sandwich assaults: a proposer’s capability to see a transaction and act on it first. At its 100 millisecond goal, oracle costs, liquidations, and on-chain auctions might replace roughly ten occasions extra typically than right this moment, a change that issues most to the DeFi protocols and merchants who depend upon that knowledge staying present.
Builders face a model of the identical query on the regulatory aspect: the Readability Act’s proposed developer protected harbor activates how a lot management over a transaction a protocol really has, the identical line Cadence is redrawing on the consensus layer.
Whether or not Cadence Hits Its 100ms Goal on Mainnet
Cadence is presently a analysis proposal backed by simulation outcomes, not a stay mainnet function. The arXiv paper, final revised July 7, 2026, units an preliminary goal block interval of 100 milliseconds, about ten occasions quicker than Monad’s present one-second block time.
Class Labs hasn’t printed a mainnet date for Cadence itself, although its personal weblog exhibits RaptorCast, one of many protocol’s supporting parts, already transport in manufacturing, suggesting additional items might roll out on an identical schedule.
What This Means for You
In case you maintain or commerce MON, Cadence would make front-running more durable, since no single proposer sees your order earlier than it lands in a block. It gained’t erase MEV, but it surely closes off the best model: one proposer seeing each pending transaction first, and quicker blocks would cut the window the place a stale worth triggers a nasty liquidation. None of that is stay but, so deal with Cadence as a design to trace, to not plan round right this moment.