Barclays, Citi, Deutsche Bank, and Standard Chartered have integrated the Falcon Time-Series Transformer (TST), a forecasting AI model built by Singapore-based Ant International, into their foreign-exchange systems.
In a statement on Thursday, Ant International said the integration comes along as it released a new version of the model that can lead a public industry benchmark. The model forecasts cash flow and foreign-exchange exposure for cross-border payments.
Ant International highlighted each bank used it within its own systems: Barclays in its BARX NetFX platform, Citi paired with its Fixed FX Rates product, Standard Chartered alongside its SCALE FX system under the Monetary Authority of Singapore’s PathFin.ai program, and Deutsche Bank in its foreign-exchange operations.
The model was first deployed internally at Ant International to manage its own cash flow and currency exposure.
The latest version, FalconTST 2.0, recorded a Mean Absolute Scaled Error (MASE) of 0.666 on a leading public benchmark for time-series foundational models. The result is at the top of the leaderboard ahead of similar models from other technology companies, with forecast accuracy above 93 percent. The benchmark claims are based on Ant International’s own testing.
Time-series models can predict continuously changing numerical data such as transaction volumes, account balances and currency positions, helping businesses judge their funding in foreign currencies. The forecasts feed into foreign-exchange hedging, where inaccurate predictions can lead companies to over- or under-hedge against currency swings.
Ant International plans to make the model a reusable forecasting tool across industries rather than a single-purpose one, such as in finance, aviation, e-commerce, and logistics.
Jiang-Ming Yang, chief innovation officer of Ant International, said the value of the model lay in turning forecasts into decisions on liquidity, currency exposure and capital allocation.
Kelvin Li, general manager of platform technology and a senior vice-president, said the improved accuracy in version 2.0 would extend the tool’s benefits to banking partners and to sectors such as e-commerce, travel and fintech.