Monad

Monad did not ask Solidity developers to declare every dependency in advance. It kept blocks as linearly ordered EVM transactions and let the client speculate. Independent work runs across cores; when a later transaction read state that an earlier transaction changed, its result is discarded and run again. Parallelism changes the schedule of computation, not the ordered answer.

MonadDb was built around that workload. It stores authenticated Merkle Patricia Trie data directly, overlaps SSD reads with other execution, and preserves versions so one writer and many readers see consistent states. Its speed has a storage boundary: recent versions are retained and older history is pruned according to disk capacity unless an operator runs separate archival infrastructure.

Asynchronous execution removes current execution from consensus's critical path. A leader proposes an ordering without the latest state root; validators vote on block validity before knowing whether each transaction will succeed. Once order is fixed, every node executes the same list. A D=3 delayed Merkle root later exposes divergence, while reserve-balance rules guard gas funding against the lagged view.

MonadBFT gives the ordering stake-weighted certificates above two-thirds. Current configuration sets 300 ms slots, one-slot speculative finality and two-slot full finality. The adjective matters: speculative finality can revert under rare leader equivocation conditions. The advertised 10,000 TPS is a designed capacity, while real throughput still depends on conflicts, transaction gas, payload limits, nodes and demand.

Mainnet and MON launched on 24 November 2025. The approximate allocation put 38.5% into Ecosystem Development, 27% Team, 19.7% Investors, 7.5% Public Sale, 3.95% Category Labs Treasury and 3.3% Airdrop. The rounded entries nearly, rather than perfectly, sum to 100%. That is a disclosure characteristic, not a missing token category.

About 49.4 billion MON was unlocked at launch. Only about 10.8 billion came from sale and airdrop public circulation; 38.5 billion was unlocked inventory stewarded by Monad Foundation for grants, incentives, expenses and validator delegation. Another 50.6 billion allocated to team, investors and Category Labs Treasury was locked for at least a year, unable to stake, with full release anticipated by Q4 2029.

Supply also became a flow rather than a ceiling. Launch blocks paid 25 MON, and base transaction fees burned. MIP-12 shortened the vote pace from 400 to 300 ms and proportionally reduced rewards. Current source activates 18 MON at round 89,758,000; the live chain was already past 101 million blocks on the review date. Repeating 25 as today's reward would turn launch literature into misinformation.

Monad staking lives in a precompile. Delegation sets validator weight by epoch, rewards can be claimed or compounded, and undelegated value waits one epoch before withdrawal. Current active admission takes at least 10 million MON plus a place in the top 200. Running the binary is open; obtaining consensus weight is a capital ranking.

Foundation's Validator Delegation Program supplied 9.3 billion MON across 170 validators. It gave authorization stake grants and at least admission-sized delegations, then graded geography, uptime, commission, security, ecosystem work and timely upgrades. The program broadens the operator count, but selection and withdrawal remain at Foundation discretion. Distribution of machines and distribution of the stake that chooses leaders are related, not identical.

Monad Foundation said it would facilitate validator-led governance and community-led proposals, while the renamed Category Labs would continue protocol research and software. MIP-1 makes that social machinery concrete: authors gather discussion, editors maintain the record, core-developer calls seek rough consensus, and a Core MIP needs an implementation before it can become Final. It describes no token-holder ballot.

Network revisions are hard forks encoded in client chain configs and scheduled for a future round or time. Operators receive official release instructions and must install compatible software; a supermajority's adoption keeps the chain moving. The July 2026 300 ms revision shows the full path from Category Labs-authored MIP to code and validator behavior.

An ordinary MON holder still has a lever. Delegating or withdrawing stake changes who may propose and certify blocks, and a validator set can accept or reject software by the clients it runs. The lever is indirect and concentrated by large delegations. It does not merge a MIP, control Foundation reserves, confer Category Labs equity or let one wallet rewrite chain parameters.