M0 is not trying to impose a new consumer-facing dollar brand. The project provides a technical stack for wallets, fintech companies and payment networks to create their own stablecoin, choose an issuer and decide how reserve income is distributed. MetaMask USD, for example, can have its own name, rules and supported chains while relying on M0 infrastructure.
This modularity addresses a real problem. Switching issuers today often means changing the token contract, liquidity and integrations as well. M0 promises to make those components replaceable. A configurable stablecoin is not an intermediary-free stablecoin, however. Assets remain offchain, the institutions holding them remain permissioned, and reserve data depends on validators. The project is most interesting at that boundary: it separates functions more cleanly without removing their risks.
What M0 actually builds
The stack has three layers. At the application layer, an “extension” defines the stablecoin’s name, access controls, presence on several chains and destination of reserve income. At the distribution layer, an orchestration service finds routes between those extensions and external stablecoins, including USDC, using orders with predetermined prices. At the issuance layer, an institution holds reserves and mints or burns units through a Stablecoin Core.
M is the historical base token for this system. The documentation describes it as an immutable ERC-20 backed one-for-one by approved collateral. Permissioned institutions called Minters place eligible assets in offchain vehicles, then register a collateral value in the `MinterGateway` contract. They may mint only a governed fraction of that value. Extensions can subsequently wrap or reuse the base to produce a branded stablecoin.
M0 also offers newer models. An issuer can operate its own core, while an extension can be built around reserves already onchain. There is therefore no single “M0” legal structure. Code standardises movement and accounting; the issuing entity, redemption claim and reserve composition vary by product.
Several brands do not mean several reserves
Portability is the main advantage. A company can retain its token and user integration while, in theory, replacing selected providers. It can also decide who receives asset income: its treasury, users or several institutional accounts. This separation avoids letting a vertically integrated issuer control the brand, reserve, contract and distribution at once.
It can nevertheless create an illusion of diversification. Two differently named stablecoins may share the same base token, issuer or liquidity route. Adding them together would then inflate activity. DeFiLlama marks M as “doublecounted” because some of its supply reappears in extensions. Analysts must distinguish the underlying supply from the user-facing tokens, then identify the legal issuer and redemption route for each.
This is the same blind spot found in the issuer groups examined yesterday: moving a function into another contract or company does not necessarily move the economic risk. Modularity makes replacement easier; it does not prove that replacement will be instant during stress or that holders will retain exactly the same rights.
Usage is real, but base supply contracted
M0 has moved beyond a prototype. Contracts, addresses, extensions and audits are public. Its website presents MoneyGram, MetaMask and KAST as case studies. DeFiLlama tracked about $183.1 million of M at approximately 06:10 UTC on 2 September 2026, down from $293.6 million one month earlier. The decline was $110.6 million, or 37.7%.
That contraction does not by itself prove users are leaving. Redemptions, migrations among forms of M or methodology changes can alter supply. It does prevent the number of integrated brands from being confused with automatic growth in the underlying money. Infrastructure can announce more partners while its monetary base shrinks.
Distribution is concentrated as well. Ethereum held $155.9 million, or 85.2% of tracked supply. Solana represented $15.5 million, Monad $9.1 million and Noble $2.0 million. All other networks together accounted for less than 0.4%. M0 therefore has genuine multichain capability, but observable activity in its base token remains overwhelmingly on Ethereum.
Marc Norat’s own analysis is to separate three indicators. The number of launched products measures commercial adoption; M supply measures the tracked common base; distribution by chain shows where that base sits. None replaces the other two, and adding them would be misleading.
Code controls rates; actors verify assets
Ethereum does not directly observe reserves. Minters periodically submit their collateral value. Several authorised validators must sign an update; they may also cancel a suspicious mint or temporarily freeze an issuer. The documentation states a maximum collateral-update interval of 30 hours. This makes attestations more frequent and usable by code, but they remain statements about offchain assets.
Governance selects Minters, validators and accounts permitted to earn yield. It also sets the mint ratio and rates. M0 uses two tokens: POWER for operating decisions and ZERO for meta-governance and a share of protocol income. POWER participation is encouraged through inflation reserved for complete voters; ZERO can intervene in fundamental changes.
The design is sophisticated but not neutral. Ordinary holders of an extension stablecoin do not necessarily vote on validators or reserves. The technical option to change issuers still requires a decision, legal migration, liquidity and operational execution. Governance reduces some forms of lock-in without turning every holder into a controller of the money.
The business model distributes reserve income
Issuers pay a rate on the supply they have created. The protocol can pass a lower or equal rate to accounts authorised to earn. The spread, penalties and selected fees feed a distribution vault linked to governance. Each extension then chooses its model: revenue to a treasury, user rewards or a split among accounts.
M0 also sells infrastructure to issuers and presents orchestration as an additional licence. The sources reviewed contain neither a public price schedule nor audited revenue. The economic channels can therefore be described, but the project’s profitability cannot be measured. More importantly, yield does not originate in code. It comes from reserve assets and issuer payments; the contracts automate its allocation.
Audits cover software, not redemption
The project publishes a substantial audit record: eight reviews of the core and governance in 2023-2024, followed by separate assessments of EVM and Solana extensions, multichain portals and the liquidity protocol through June 2026. That depth of documentation is a positive signal.
An independent 2024 core review by Kirill Fedoseev found one high-severity signature double-counting issue, marked fixed, alongside acknowledged observations. Its broad warning remains useful: security ultimately depends on governance incentives and the presence of at least one honest validator. This does not prove a current vulnerability. It identifies the structural boundary of a system connecting software to offchain guarantees.
Newer versions add further surfaces: administrable extension contracts, liquidity solvers, bridges and messaging adapters. Each has been audited within a defined scope, but no software report certifies issuer solvency, legal segregation of reserves or the speed of a bank redemption.
What to verify next
M0 provides accessible, used and technically differentiated infrastructure. It can reduce launch costs and dependence on one permanent issuer. Its decisive test is still economic: an issuer change must work in production without breaking liquidity, rights or the peg.
Four disclosures would make the project easier to assess: non-double-counted supply for every extension; the identity, jurisdiction and reserve reports of every issuer; actual conversion volumes and settlement times into bank money; and concentration among POWER votes, ZERO holders and validator signatures. Until then, M0 is best understood as a coordination layer for several stablecoins, not as automatic diversification of their reserves.