Monad is a layer 1 blockchain that runs Ethereum smart contracts without modification while executing many of them at the same time, targeting 10,000 transactions per second with 800 millisecond finality. Its native token, MON, pays gas, secures the chain through delegated staking, and has a fixed initial supply of 100 billion. Mainnet went live on 24 November 2025 after roughly three years of engineering and close to half a billion dollars in venture and public sale funding.
The pitch is narrow and unusually testable. Every other high-throughput chain asked developers to learn something new. Monad asked them to change nothing and re-engineered the four layers beneath the virtual machine instead.
The Bottleneck Monad Was Built to Remove
The Ethereum Virtual Machine processes transactions one after another. Each one reads state, writes state, and only then does the next one begin. That design makes correctness easy to reason about and makes throughput a function of a single CPU core plus disk latency, which is why Ethereum settles in the low tens of transactions per second and why gas spikes when demand arrives in bursts.
Two schools of thought emerged in response. One replaced the virtual machine outright: Solana built Sealevel, the Move chains built new languages, and each asked developers to rewrite their contracts and their tooling. The other kept the EVM and moved execution off-chain into rollups, inheriting Ethereum's security at the cost of fragmented liquidity and withdrawal delays.
Monad took the third path. Ethereum bytecode runs byte for byte identically, the RPC interface matches, and the same Solidity, Foundry, Hardhat, and wallet tooling works unchanged. The performance came from rebuilding consensus, networking, execution scheduling, and state storage as four separate engineering problems.
The Five Parts of Monad's Performance Stack
Monad's throughput claims rest on five components that each attack a different constraint. The project's technical documentation covers all five in depth.
MonadBFT
Consensus derives from HotStuff, with each validator sending its vote to the block leader, who aggregates them. Communication scales linearly with validator count rather than quadratically, which keeps the messaging overhead flat as the set grows. Blocks arrive roughly every 400 milliseconds and reach finality in about 800 milliseconds, so a transaction is irreversible in under a second.
RaptorCast
A block that no one has received cannot be voted on, so propagation is a real ceiling on block size. RaptorCast splits each block into erasure-coded chunks and pushes them through a two-level broadcast tree. Any validator can rebuild the full block from a subset of chunks roughly matching the original block size, which means one slow or offline peer does not stall the round.
Asynchronous Execution
Ordering and execution are decoupled. Validators agree on the sequence of transactions in a block and finalize it, then execute those transactions in a separate pipeline while consensus has already moved on to the next block. Execution gets an entire block interval to itself instead of competing with voting for the same wall clock.
Optimistic Parallel Execution
Transactions still commit in their canonical order, but the engine runs them concurrently across CPU cores on the assumption that most will touch unrelated state. It tracks which slots each transaction read and wrote, and when two conflict, the later one is discarded and re-executed against the corrected state. Blocks of independent swaps, transfers, and mints finish in a fraction of the time sequential execution would need.
MonadDB
The final constraint is disk. MonadDB stores the Merkle trie natively on SSD rather than layering it on a general-purpose key-value store, and issues state reads asynchronously so thousands can be in flight at once. Without it, parallel execution would simply queue up behind I/O.
MON Supply and What It Pays For
MON launched with a total supply of 100 billion tokens. The Monad Foundation's tokenomics disclosure breaks the initial allocation down as follows.
| Allocation | Share | Purpose |
|---|---|---|
| Ecosystem development | 38.5% | Grants, incentives, liquidity programs |
| Team and core contributors | 27.0% | Locked one year, then vests over three |
| Investors | 19.7% | Four-year schedule, one-year cliff, monthly after |
| Public sale | 7.5% | Sold at $0.025 to 85,820 participants |
| Treasury | 4.0% | Foundation operations and reserves |
| Community airdrop | 3.3% | Claimed by roughly 230,000 wallets |
About 50.6% of supply was locked at launch and roughly 10.8% circulated, according to The Block's coverage of the launch. Circulating supply reached 11.83 billion MON by August 2026. Staking rewards mint roughly 2 billion MON a year on top of that, an inflation rate near 2% that funds validator and delegator yield.
The date that matters most sits in November 2026, when the one-year cliff on team and investor allocations expires and monthly unlocks begin running through 2029. Close to 47 billion tokens sit behind that cliff.
What Runs on Monad Today
Total value locked reached roughly $408 million by May 2026, spread across lending markets, DEXs, and early real-world asset experiments. The composition is more interesting than the headline number, because most of it clusters in applications that only make sense on a fast chain.
Kuru operates a fully on-chain central limit order book with no off-chain sequencer and no custodial matching engine, a design that is impractical when every order placement and cancellation costs a full block of latency. aPriori pairs liquid staking with a probabilistic MEV auction, issuing aprMON that accrues both staking yield and redistributed MEV. Magma takes a different approach to the same problem, using distributed validator technology to back its gMON liquid staking token across multiple operators. Curvance handles modular liquidity management on top.
Established protocols arrived too. Aave's deployment brought a rented liquidity model to Monad, importing depth rather than bootstrapping it from zero. On the consumer side, Fantasy Top brought its crypto-personality trading card game over from an Ethereum layer 2, joining a gaming and NFT cohort that Monad courted through hackathons well before mainnet existed.
What MON Exposure Actually Tracks
Buying MON is a bet on adoption of parallel execution rather than a bet on whether it works. The engineering already demonstrated itself on testnet and in production. The open question is whether developers and users route enough volume through Monad to make 10,000 TPS a constraint anyone bumps into.
Three variables drive that. Fee revenue scales with real usage, and gas on a chain this fast is close to free, so the token needs enormous transaction counts before fees matter to holders. Staking demand locks supply away and pays roughly 13% at current participation, which competes directly against the temptation to sell. And the November 2026 unlock schedule introduces a supply curve that has to be absorbed by demand that did not exist at launch.
MON traded around $0.03 in August 2026 with a market capitalization near $355 million, well below the $0.049 all-time high it printed two days after mainnet and below its $0.025 public sale price for stretches of the year. Traders who want directional exposure without holding spot can use margin and leverage on the perpetual contract, though a token with a known unlock cliff ahead of it deserves position sizing that assumes volatility rather than hoping against it.
Common Questions About Monad
What is Monad used for?
Monad is used to deploy and run Ethereum smart contracts at higher throughput and lower cost than Ethereum itself. Developers use it for DeFi protocols, order-book exchanges, games, and NFT applications that need sub-second confirmation. Users pay gas in MON and can stake it to help secure the network.
Is Monad a layer 2 on Ethereum?
Monad is a standalone layer 1 blockchain with its own validator set and its own consensus. It shares Ethereum's execution environment and developer tooling, while settling its own transactions rather than posting them to Ethereum for security the way a layer 2 does.
How many transactions per second can Monad handle?
Monad targets 10,000 transactions per second with 400 millisecond blocks and 800 millisecond finality. That figure comes from testnet benchmarking and the theoretical capacity of its parallel execution engine, and real throughput on mainnet tracks actual demand, which remains far below the ceiling.
What is the total supply of MON?
MON has a total supply of 100 billion tokens, of which roughly 11.83 billion circulated as of August 2026. Staking rewards add approximately 2 billion MON per year, so supply grows at about 2% annually on top of scheduled unlocks.
Can you stake MON?
Yes. Monad uses delegated proof of stake, so holders can delegate any amount of MON to a validator without giving up ownership, and there is no minimum. Rewards accrue each epoch, run near 13% annualized, and are claimed or restaked manually rather than compounding automatically.
Where Monad Goes From Here
Monad answered a question the industry had been circling for years: whether the EVM's throughput ceiling was a property of the instruction set or a property of everything built around it. The answer turned out to be the latter, and the proof is a chain where unmodified Solidity settles in under a second.
The harder question is commercial. Monad enters a market with mature high-performance alternatives on one side and Ethereum's entrenched liquidity on the other, and its window to convert technical credibility into durable activity narrows as the 2026 unlock schedule arrives. Watching TVL composition, daily active addresses, and staking participation gives a clearer read on that than watching price alone.
Ready to take a position? MON spot pairs are live on LeveX, MON perpetuals offer leveraged exposure in both directions, and Crypto in a Minute covers the rest of the market in the same format.