Project Agorá Settles About $1 Million Across Six Currencies in Live Tokenised Deposit Tests
Oct. 26, 2026 • Deep Dive • Digital Assets
Project Agorá settled approximately CHF 800,000 in live, multi-currency transactions during a July 2026 testing phase. The work examined a cross-border payment model that places tokenised commercial bank deposits on a shared ledger while holding tokenised central bank reserves on jurisdictional ledgers. Five central banks participated directly in live-reserve settlement, although the project’s wider central-bank group has since expanded to eight with the Bank of Canada’s entry.
The project reports an average settlement time of 80 seconds across six currencies and uses smart contracts to run anti-money laundering, sanctions and fraud checks simultaneously. Legal analyses found that tokenisation did not change the underlying character of deposits or reserves in participating jurisdictions. The findings support further work on deposit-based wholesale settlement, but the supplied research does not establish that the approach outperforms stablecoins. The Eurosystem plans to draw on Agorá’s architecture for Projects Pontes and Appia.
Project Agorá settled approximately CHF 800,000, or roughly $1 million, in live multi-currency transactions during a July 2026 testing phase. The result moves a model based on tokenised commercial bank deposits beyond the prototype findings reported in May, while leaving open the question of whether it performs better than stablecoins in wholesale cross-border settlement.
The project is testing how banks could transfer deposits across borders using a shared programmable ledger alongside tokenised central bank reserves. Its reported average settlement time was 80 seconds across six currencies: the US dollar, euro, British pound, Japanese yen, Swiss franc and South Korean won. The supplied research does not identify a comparable stablecoin test, so the timing cannot establish which model is faster.
What the live testing established
Five central banks participated directly in the July real-value testing: the Bank of England, Bank of France, Bank of Japan, Bank of Korea and Swiss National Bank. The Federal Reserve Bank of New York and Bank of Mexico were part of the original seven-central-bank group but did not execute live-reserve settlement transactions in that phase. The Bank of Canada has since joined Project Agorá as an eighth participating central bank.
That distinction matters when assessing the test’s scope. The project involved six fiat currencies, but participation in the wider initiative should not be confused with a central bank settling a live transaction using its reserves. The reported CHF 800,000 is an aggregate value for the live multi-currency activity, not evidence of deployment at commercial scale.
The supplied research describes the July workload in differing ways, referring to both transaction scenarios and individual transactions. Those counts are not directly comparable, so the aggregate settlement value and participating currencies provide the clearer account of what was tested.
How the deposit-based design works
Project Agorá uses a two-layer programmable ledger architecture. Tokenised commercial bank deposits sit on a shared unifying ledger, while tokenised central bank reserves are held on jurisdictional ledgers. Tokenisation represents an existing financial claim on a ledger; it does not, by itself, turn a bank deposit into a different kind of money.
The design aims to retain the two-tier banking system, in which customers hold claims on commercial banks and banks settle using central bank reserves. That distinction is central to the case for tokenised deposits rather than private stablecoins. Central banks prioritise the deposit approach because it preserves that structure and the intended singleness of money across commercial and central bank money.
The platform also allows anti-money laundering, sanctions and fraud checks to run simultaneously through smart contracts, rather than sequentially across correspondent intermediaries. That changes the proposed sequence of payment processing; the live test alone does not demonstrate how the model would perform across every bank, jurisdiction or payment flow.
Legal and regulatory analyses across the participating jurisdictions found that tokenisation did not alter the underlying legal character of commercial bank deposits or central bank reserves, and that settlement finality was valid. Those findings address a prerequisite for using ledger-based representations of existing money, alongside the technical question of whether the transactions can be processed.
Enterprise blockchain software firm Kaleido supplied the underlying technology platform for the Project Agorá prototype.
What remains to be proved
The potential market is substantial: wholesale cross-border payment flows were estimated at $195 trillion in 2024, representing roughly 91% of global cross-border payment value. The July test, however, settled a much smaller amount. It provides evidence about the feasibility of live settlement under the tested arrangements, not a measure of how the platform would handle that broader volume.
Nor does the research provide a like-for-like comparison with stablecoins on cost, speed, liquidity requirements or operational resilience. Agorá’s reported settlement time and legal findings make a case for further testing of tokenised deposits, but they do not show that deposits outperform another payment instrument.
The Eurosystem has said architectural insights from Agorá will inform Project Pontes, which is scheduled to connect distributed ledger platforms to TARGET Services, and Project Appia, which outlines a longer-term tokenised financial ecosystem for Europe. Those plans show where the work may feed into future infrastructure. They do not amount to a timetable for broad commercial use of the Agorá model.
Ramp partnered with Stripe and Bridge in July 2026 to deploy stablecoin rails for business bill payments and yield-bearing stablecoin accounts funded through standard fiat bank accounts. The arrangement shows how stablecoins can be incorporated into familiar payment workflows without requiring businesses to fund every transaction with digital assets. The economics depend on more than the cost of moving a token on-chain. Stripe charges merchants 1.5% for stablecoin checkout transactions, versus its standard card-processing fee of 2.9% plus $0.30; neither figure establishes the price of Ramp’s bill-payment service. Businesses receiving fiat may also face foreign-exchange and off-ramp costs, typically 0.3% to 2.0% when converting into local banking systems. Domestic clearing networks and compliance checks remain part of end-to-end delivery. Bridge’s issuance and transfer tools, alongside its conditional U.S. trust-bank approval, illustrate the infrastructure behind those payments—but not the elimination of every payment cost.
StraitsX has opened onboarding for digital payment token providers across more than 100 markets through an expanded card-issuing architecture developed with Visa. Integration documentation published on 26 October 2026 describes a multi-BIN model that combines issuing sponsorship, transaction authorization, compliance screening and fiat clearing. The approach lets participating providers build stablecoin-backed card products while merchants continue to receive conventional fiat payments through Visa’s network. StraitsX says its card gross transaction value rose 40-fold year on year from Q4 2024 to Q4 2025, while cards issued across Asia increased 83-fold. On-chain issuer telemetry recorded monthly card transaction volume of $310 million to $438 million through October 2026. The company’s stated $6 billion annualized card transaction figure is distinct from its cumulative $70 billion in processed stablecoin volume. StraitsX and Visa have also scheduled demonstrations of agentic payments for November 2026.
ACI Worldwide and Circle announced a pilot on October 27, 2026, involving USDC and connectivity to Colombia’s Bre-B instant payments system. The integration runs through ACI Connetic, a payments orchestration platform, rather than embedding USDC in Bre-B’s central infrastructure. There is no indication that Colombia’s central bank has adopted USDC as a settlement asset for the national network. The distinction matters because ACI also supplies technology for Bre-B’s central directory and real-time clearing engine. Bre-B launched in October 2025 and handled more than 500 million transactions within five months. Connetic is designed to bring traditional payment rails and digital asset networks together without requiring banks to replace legacy core ledgers. ACI says the platform can support other digital assets and liquidity providers, so the pilot does not make USDC its exclusive option. The arrangement tests a vendor-level connection alongside an established public payment system, not a change to that system’s settlement rules.
Ripple received final approval on 26 October 2026 for Ripple National Trust Bank, giving it a federally chartered institution for fiduciary and custody activities. The charter does not authorize commercial deposits or loans, and it does not transfer issuance of Ripple’s RLUSD stablecoin to the new bank. RLUSD issuance remains with Standard Custody & Trust Company under New York State Department of Financial Services supervision. BNY is the primary custodian of the dollar deposits and U.S. Treasury reserves backing the stablecoin. The arrangement separates a federal custody role from state-supervised issuance. It also comes as the OCC implements the GENIUS Act’s stablecoin rules and faces a lawsuit from the Independent Community Bankers of America challenging its authority to grant national trust charters to crypto firms. Ripple has separately announced an expanded partnership with BNY and Robinhood focused on institutional custody capabilities and exploring tokenized trade finance.
Stripe confirmed in October 2026 that it plans to expand its Bridge- and Visa-powered stablecoin card issuing program from 18 initial countries to more than 100 by the end of the year. The planned rollout spans Europe, Asia-Pacific, Africa and the Middle East. It follows Stripe’s February 2025 acquisition of Bridge, whose APIs connect stablecoin transfers with fiat accounts and bank payouts. Stripe has also made Open USD available to businesses in Stripe Financial Accounts alongside USDC. Its broader stablecoin infrastructure combines Bridge’s conversion tools with Stripe’s existing payment and treasury network and embedded wallets from Privy. Lead Bank supports the card program as a financial institution partner and BIN sponsor. A separate deployment by Ramp uses Bridge APIs and Privy wallets to let businesses pay vendors in stablecoins using funds from fiat bank accounts. Together, the developments show how Stripe is assembling stablecoin services for business payments, while the planned geographic expansion remains unfinished.
Project Agorá settled approximately CHF 800,000 in live, multi-currency transactions during a July 2026 testing phase. The work examined a cross-border payment model that places tokenised commercial bank deposits on a shared ledger while holding tokenised central bank reserves on jurisdictional ledgers. Five central banks participated directly in live-reserve settlement, although the project’s wider central-bank group has since expanded to eight with the Bank of Canada’s entry. The project reports an average settlement time of 80 seconds across six currencies and uses smart contracts to run anti-money laundering, sanctions and fraud checks simultaneously. Legal analyses found that tokenisation did not change the underlying character of deposits or reserves in participating jurisdictions. The findings support further work on deposit-based wholesale settlement, but the supplied research does not establish that the approach outperforms stablecoins. The Eurosystem plans to draw on Agorá’s architecture for Projects Pontes and Appia.