The model, called FiMI, short for Finance Model for India, was released at the India–AI Impact Summit 2026 in New Delhi. Developed in-house, FiMI currently powers NPCI’s UPI Help Assistant, an AI-driven conversational support system for UPI users.
NPCI said the model is designed to address “limitations observed in general-purpose large language models when applied to high-scale, high-trust payment workflows.”
FiMI has been built to understand Indian payment systems, including UPI, and covers areas such as transaction dispute handling, mandate lifecycle management, and regulatory and ecosystem queries. The model underwent continuous pre-training and was further fine-tuned on Indian financial and synthetically generated payments data, NPCI said.
The payments body said the model enables “accurate reasoning, structured tool invocation, and robust multilingual support,” adding that its “payments-native design allows FiMI to operate reliably in real-world, high-volume transaction environments where accuracy, consistency, and trust are critical.”
The UPI Help Assistant, deployed as a pilot initiative at a national scale, uses an agentic AI framework that supports multi-step reasoning to respond to payment-related queries, assist with grievance redressal, and manage mandates. The assistant currently supports English, Hindi, Telugu, and Bengali, with additional Indian languages expected within six to eight months.
The organisation said it is exploring next-generation model architectures, including Mixture-of-Experts, and expanding multilingual capabilities to cover more Indian languages. These initiatives remain in the research and experimentation phase and are aimed at improving scalability, efficiency, and domain intelligence, while maintaining governance and reliability standards required for national-scale payment systems.
NPCI said it has published a detailed technical paper on the arXiv platform outlining FiMI’s data curation process, training methodology, and evaluation results.
“FiMI reflects NPCI’s commitment in developing financial language models purpose-built for India’s payments ecosystem. Having successfully built and validated the model, we are now sharing our technical learnings through a detailed paper to contribute to responsible innovation, while strengthening trust, transparency, and collaboration across the ecosystem,” Vishal Kanvaty, Chief Technology Officer at NPCI, said.
Edited by Jyoti Narayan
Discover more from News Link360
Subscribe to get the latest posts sent to your email.
