Project mBridge: where it landed

Project mBridge set out to prove that central banks could settle cross-border payments directly in their own digital currencies, on one shared platform, without correspondent chains. It proved that, technically. It reached minimum viable product in mid-2024, settled real-value pilot transactions, and became the most advanced multilateral central bank digital currency platform ever built. Then its convenor stepped back, and the project’s story became a lesson in how settlement infrastructure actually evolves: toward regional venues that do not connect, with the one-global-platform vision receding as they multiply.

What mBridge built

The platform is a shared ledger on which participating central banks issue their own wholesale digital currencies and commercial banks in their jurisdictions settle cross-border payments in them, payment versus payment, in seconds. The founding members were the central banks of China, Hong Kong, Thailand, and the UAE; the Saudi Central Bank joined as a full member in June 2024, and around 32 institutions observe, including the IMF and the World Bank.

Technically, the BIS’s own page describes a platform compatible with the Ethereum Virtual Machine, built as a testbed for new use cases and for interoperability with other platforms. A payment between a bank in Hong Kong and a supplier’s bank in the UAE crosses one ledger in one step, with no nostro account chain, no pre-funding of positions along the way, and finality in central bank money.

The handover

In October 2024 the BIS announced it was handing mBridge to its partner central banks, moving itself to observer status. The official framing was graduation: the BIS Innovation Hub incubates projects and exits once they can stand alone, and minimum viable product was the agreed milestone. The BIS General Manager was also candid that the platform was not mature enough to start operating.

The timing invited a second reading, which serious commentary has aired and which is worth stating factually: the exit came shortly after a BRICS summit at which cross-border payment alternatives were on the agenda, mBridge’s remaining members all sit outside the G7, and the platform is widely read as a route for trade settlement that runs outside the dollar correspondent system. In the same period, the BIS convened Project Agorá with a membership centred on reserve-currency central banks. Whatever the weighting of reasons, the outcome is plain: the official sector’s two flagship settlement experiments now sit on opposite sides of a geopolitical line, with no bridge between them.

Where it stands in 2026

The BIS’s careful phrase is that the platform is enabled to undertake real-value transactions, subject to jurisdictional preparedness. The member central banks have said they are advancing mBridge, and analysts describe its practical role today as a settlement rail for trade between China and the Gulf. A multilateral production launch, with published volumes and open membership, has not been announced. For corridor coverage, the boundary is set by its membership: renminbi, Hong Kong dollar, baht, dirham, and riyal flows are in scope; dollar, euro, and sterling flows are not.

The lesson for settlement

mBridge matters less as a service than as evidence, on three points. Cross-border settlement in central bank money, in seconds, is achievable; the technology was never the barrier. Governance, not code, decides who can use a settlement platform; membership is the product. And the future is plural: mBridge for one set of corridors, Agorá’s eventual platform for another, bank consortium ledgers, single-bank networks, and stablecoin rails alongside, each partial, each disconnected. An institution planning its settlement stack around any single venue is betting on a map that keeps redrawing itself.

Where Frame fits

Frame is the settlement layer for global finance: one integration to orchestrate payments at scale across fiat rails, stablecoins, and tokenized deposits, with compliance enforced on every transaction and settlement in seconds instead of days.

mBridge is exactly the kind of development a rail-neutral settlement layer is built for. Frame routes each payment across whichever rail fits the corridor, the counterparty, and the policy that governs it, so an institution’s answer to a fragmenting map is a policy question, which rail serves this flow best today, rather than a commitment to one venue. Compliance is enforced inside settlement through Frame’s Rules Engine, and every settled transaction produces verifiable evidence that its conditions were met, whichever rail carried it. If mBridge corridors open to more institutions, that is one more rail to span, and the payments that cannot use it still settle in seconds on the rails that exist.

See how a rail-neutral settlement layer works: the Frame Blueprint.