This part is the operator's reference. It walks through the production and advanced-pilot atomic delivery-versus-payment (DvP) deployments a 2026 reader is most likely to meet, ordered roughly from the cleanest single-bank case through to the most ambitious multi-currency multi-jurisdiction work. Each example is here for what it teaches structurally, not for league-table reasons. Skim the ones that are not relevant to your patch and read the ones that are with the chapter's framing in mind: which CPMI model, what cash leg, which atomicity pattern.
The Kinexys worked example
Kinexys-routed tokenised deposit plus tokenised security DvP is the worked example of the Quorum-lineage approach. JPMorgan issues a tokenised deposit against its own balance sheet, the asset issuer mints on the same chain or on a linked chain, and the DvP commits as a single ledger entry (Kinexys overview). The atomic property holds within the Kinexys perimeter; cross-perimeter movements rely on coordinator-pattern bridges to other regulated venues.
The published Onyx tokenised collateral case study (Onyx is the legacy brand, Kinexys is the current one) is the only operational disclosure with named counterparties and timestamps and is the cleanest single reference for what an institutional intraday DvP product actually looks like in flight. The product fits the operator-pattern-meets-same-ledger-atomicity case, and the residual cross-perimeter movements are exactly where the coordinator pattern survives in production.
Partior cross-currency PvP
The DBS, JPMorgan, and Standard Chartered Partior arrangement settles cross-currency interbank flows on a shared Quorum-derived ledger. Each participating bank represents its deposit liabilities on the chain in the relevant currency, and a cross-currency payment-versus-payment (PvP) commits atomically against both currency legs. This is the cleanest production example of multi-bank cross-currency atomic settlement among GSIBs in APAC and the structural reference for any team designing a similar interbank arrangement. The Partior case is also instructive because it is multi-bank without a central bank settlement layer, which sets up a useful contrast with Project Ensemble and Project Agorá.
Project Ensemble
Project Ensemble in Hong Kong, run by HKMA, is the regulator-coordinated example of the tiered-ledger architecture (HKMA Ensemble overview). Commercial banks issue tokenised deposits at the deposit layer, asset issuers mint at the asset layer, and the wholesale CBDC (wCBDC) sits at the settlement layer. Atomic DvP across the asset and cash legs is enforced at the chain level, with finality flowing through the designated-system rulebook.
The HKMA's Phase 2 announcement extended the sandbox into more asset classes and more participating banks and is the live operational reference. Ensemble is the most regulator-forward of the live programmes and the natural one to point at when explaining what a tiered ledger architecture actually looks like rather than what one is supposed to look like.
Project Mariana
Project Mariana, run by the BIS Innovation Hub with the central banks of France, Singapore, and Switzerland, tested cross-border atomic settlement of tokenised wholesale CBDC across multiple currencies using an automated market maker (AMM) model. The interest is less the AMM mechanism (a trader's tool ported into wholesale settlement) and more the demonstration that cross-border, cross-currency atomic settlement is achievable in a regulated context with central bank money on each leg. Mariana is the cleanest published example of a unified-settlement-layer cross-currency design, even though no central bank has yet committed to a production AMM-on-wCBDC system.
Project Agorá
Project Agorá takes the same architectural intuition and extends it to multi-currency tokenised commercial bank money under central bank coordination. Design choices around atomicity, governance, and access are still being worked through across the participating central banks (including BIS, Bank of Japan, Bank of Korea, and others). The headline atomicity question is whether the system commits cross-currency legs simultaneously across multiple central bank systems or relies on a coordinator at the BIS layer; both designs have been on the table.
Agorá is also the single most useful reference for tokenised correspondent banking, because it is the first regulator-coordinated work that takes the existing correspondent banking stack as the starting point and asks how tokenisation reshapes it, rather than treating correspondent banking as a problem to be replaced.
Eurex repo and the BNY tokenised collateral programme
The Eurex repo work in Europe and the BNY Mellon (BNY) tokenised collateral programme in the United States are the buy-side and triparty counterparts. Tokenised collateral can be DvP'd against tokenised cash in intraday repo, releasing capital faster and supporting a more granular intraday liquidity product than legacy infrastructure permits. The institutional appetite has been particularly strong on intraday repo because the savings show up in real basis points on funding cost. The Eurex repo platform is the European reference point and the BNY programme is the US one; both are the natural counterweights to Kinexys for a reader who wants to see how the buy-side and the post-trade infrastructure pieces fit together rather than the dealer-bank piece.
The April 2026 Asian JGB pilot
The April 2026 Nomura, Mizuho, JSCC tokenised JGB collateral trial on Canton Network with Digital Asset is the most recent named pilot in japan, with two of the three previously announced as participants in the FSA Japan-supervised Payment Innovation Project sandbox. The pilot is structurally interesting because it brings the central clearing counterparty (JSCC) onto the chain alongside the dealer banks, which begins to address how atomic DvP coexists with a CCP-cleared market. The CCP question, raised in earlier parts as independent of DvP in principle, becomes practical when a regulated CCP is one of the chain participants rather than a bystander. Watch this trial for what it surfaces about CCP integration on Canton, because it is the cleanest live test of a question that the rest of the industry has so far been able to defer.
What to read next
Chapter VII on tokenised deposits is the natural next step, because atomic DvP without a credible cash leg is theatre, and tokenised deposits are the dominant cash-leg option in production today. Chapter VIII on wholesale CBDC covers the central-bank-money path and the BIS coordination work that ties atomic DvP architectures across jurisdictions.
The wiki page for Project Ensemble is the operational deep-dive on the HKMA-led tiered-ledger architecture. The page for Canton Network covers the cross-domain transaction model underwriting several of the named pilots above. The thematic page on Tokenised collateral tracks the named pilots in the bond and repo space.