Monad went live on 24 November 2025, and it arrived with more credibility than most launches manage. Three former Jump Trading engineers — Keone Hon, James Hunsaker and Eunice Giarta — spent three years rebuilding the parts of an Ethereum client that have always been the bottleneck, and raised $225m from Paradigm and Electric Capital to do it. A further $269m came in through a token sale on Coinbase to 85,820 buyers.
Nine months on, the technology works and — contrary to the usual pattern for new chains — people are actually using it. The problem is somewhere else entirely.
What it actually does
The EVM executes transactions one at a time, in order, because any transaction might touch state that a previous one changed. That sequential constraint is the ceiling on Ethereum throughput, and it has nothing to do with block size or gas limits.
Monad's answer is speculative parallel execution. It assumes most transactions in a block do not touch the same state, runs them simultaneously, then checks the assumption and re-runs any that conflicted. Most of the time the optimism is justified — a thousand people swapping different token pairs genuinely do not collide.
Around that sit three more rewrites: MonadBFT, a pipelined consensus protocol; MonadDB, a state database built for blockchain access patterns rather than borrowed from general-purpose software; and a reworked mempool. The claim is 10,000 TPS at roughly 400ms finality, with full bytecode compatibility — any Solidity contract compiled for Ethereum runs unchanged.
The pitch is Solana's throughput without Solana's toolchain. Existing contracts deploy without a rewrite, and MetaMask works on day one.
The number that changed my mind
Total value locked sits around $889m, which for a nine-month-old chain is a strong result. Uniswap and Morpho deployed within the first week, and roughly 120 protocols are now live.
The usual complaint about new chains is that TVL is rented — mercenary capital chasing incentives, generating no real economic activity. The test for that is fees, because a chain can buy deposits but nobody pays transaction fees for fun.
Monad is currently taking around $198,000 a day in fees, roughly $1.5m over the past week. For scale, Arbitrum — an established rollup with years of history — took about $217,000 over the same 24 hours. Monad is doing comparable fee volume on a fraction of the track record.
That is not a chain waiting for a use case. It is a chain being used, which is a materially different position from where most 2026 launches sit at nine months.
Set against Ethereum's $8.95m and Solana's $8m a day, Monad is still two orders of magnitude away from the leaders. But the comparison that matters at this stage is not Ethereum. It is the cohort it launched alongside, and against that cohort it is doing well.
The tokenomics problem
MON has a 100 billion token supply, of which 50.6% was locked at launch across team, investors and treasury, with quarterly unlocks running from mid-2026 through 2029. Only 10.8% was circulating on day one.
This is the structure Arthur Hayes and others have criticised, and the criticism is arithmetic rather than opinion. A token with 10% of supply circulating and nine years of scheduled unlocks ahead has a persistent, calendarised seller on one side of the book. April's unlock of 170 million MON was followed by the usual pattern of weakness as recipients distributed into whatever liquidity existed.
MON currently trades around £0.0156, giving a market cap near £184m — against a fully diluted figure many times that. Anyone buying today is buying ahead of years of supply they cannot outbid.
The harder question
Monad is a genuinely impressive piece of systems engineering aimed at a problem that was largely solved while it was being built. When work started in 2022, Ethereum congestion was the defining constraint in the industry and $50 gas fees were routine. By the time mainnet shipped, rollups had absorbed most of that demand, Base and Arbitrum were processing cheap transactions at scale, and Solana had spent two years demonstrating high throughput was achievable.
The fee numbers suggest that did not matter as much as it should have. Monad found users anyway. But "found users" and "found users who will stay when the incentives stop" are different claims, and only the first is currently evidenced.
Monad is one of several 2026 launches making different bets on the same thesis: MegaETH on real-time L2 sequencing, Plasma on stablecoin-native transfers, Converge on institutional RWAs. Each is betting the EVM is the right developer surface and its runtime is the wrong engine. They cannot all be right, and the market has not yet shown it needs any of them.
What would change the picture
Fees holding up as incentive programmes taper, which would confirm the activity is organic rather than bought. TVL surviving a major unlock intact. And an application that could not exist elsewhere — a fully on-chain order book, something that genuinely needs 400ms finality — rather than the same DeFi primitives redeployed on faster rails.
The usage question has been answered better than expected. The supply question has not been answered at all, and it is the one that will decide what MON is worth in 2029.
TVL and fee figures from DefiLlama; MON price and market cap from CoinGecko; all retrieved 14 August 2026. Launch, funding and supply figures from public reporting and project disclosures. Nothing here is a recommendation — MON is a highly volatile asset with substantial scheduled supply increases ahead.