Ant International, a global digital payment, digitization, and financial technology provider, has released its proprietary Falcon TST (Time-Series Transformer) AI model, which is now available for open-sourcing.
This industry-first Mixture of Experts architecture-based big data model utilizes multiple patch tokenizers and is supported by up to 2.5 billion parameters. Falcon TST has achieved state-of-the-art zero-shot results on long-term forecasting benchmarks, particularly in absolute mean error rates.
Internally, Ant International has deployed Falcon TST to manage cashflow and foreign exchange (FX) exposure on an hourly, daily, and weekly basis. The model has demonstrated accuracy rates exceeding 90% and has contributed to a reduction in the company’s FX costs by up to 60%.
Ant International has also collaborated with various industry partners to implement the model in different use cases. These applications include assisting businesses in mitigating FX cost and volatility, and supporting airlines in offering more stable and competitive pricing to their customers. With Airports Council International World forecasting global air travel to reach nearly 10 billion passengers in 2025, these AI applications have the potential for significant cost savings for consumers globally.
Beyond FX management, the Falcon TST model enables businesses to make other time-series forecasts, such as weather patterns, calendar events, financial market fluctuations, and cross-border traffic data.
Jiang-Ming Yang, Chief Innovation Officer at Ant International, stated, “By open-sourcing our proven Falcon TST model, we aim to advance the field through global collaboration — inviting scientists worldwide to contribute real-world feedback and accelerate innovation in time series learning.”
The Falcon TST AI Model is accessible on GitHub and Hugging Face, where interested developers can find more information about the model and its recent success stories.
Ant International, headquartered in Singapore with operations across Asia, Europe, the Middle East, and Latin America, provides a unified techfin platform to support financial institutions and merchants with digital payment and financial services solutions.
