Barclays Bank will utilize Ant International’s proprietary Time-Series Transformer (TST) AI FX Model to help businesses mitigate FX-related costs and risks arising from global volatility.
Ant International’s TST Model is a transformer architecture-based big data model featuring nearly 2 billion parameters. By integrating advanced time series forecasting algorithms, it predicts patterns over time. Additionally, Ant has developed new pre-training and Supervised Fine-Tuning (SFT) frameworks to refine the model’s predictions continuously.
The collaboration has already seen success, with Ant International completing its first batch of intra-group FX transactions with Barclays. The TST Model now forecasts the company’s cash flow and FX exposure on an hourly, daily, and weekly basis with over 90% accuracy. This enhancement allows Barclays to lower its hedging costs, enabling Ant to offer more competitive rates to business clients.
Barclays has integrated the TST Model into its FX hedging platform, BARX NetFX, which primarily serves the e-commerce and payment sectors.
Ben Parkinson, head of Global Fintech & FX Automation Sales at Barclays, expressed enthusiasm about the partnership: “Ant International has been a valued and long-standing partner of Barclays. Their cutting-edge AI model has improved the accuracy of forecasting cash flows and optimized the FX hedging process. By combining Ant’s advanced capabilities with our market-leading FX expertise, we’ve reduced uncertainty and costs, setting a new standard for FX risk management.”
Currently, the use case supports major currency pairings utilized by Ant International, but both companies plan to expand the product to cover additional currencies and address broader business needs.
Pushkaraj Gumaste, head of Corporate Banking for Asia Pacific & Middle East at Barclays, noted: “This collaboration reflects Barclays’ commitment to evolve alongside our partners, leveraging our complementary strengths to enhance our offerings and deliver impactful solutions.”