FiMI: A Domain-Specific Language Model for Indian Finance Ecosystem
Abstract
We present FiMI (Finance Model for India), a domain-specialized financial language model developed by National Payments Corporation of India (NPCI) for Indian digital payment systems. We develop two model variants: FiMI Base and FiMI Instruct. FiMI adapts the Mistral Small 24B architecture through a multi-stage training pipeline, beginning with continuous pre-training on 68 Billion tokens of curated financial, multilingual (English, Hindi, Hinglish), and synthetic data. This is followed by instruction fine-tuning and domain-specific supervised fine-tuning focused on multi-turn, tool-driven conversations that model real-world workflows, such as transaction disputes and mandate lifecycle management. Evaluations reveal that FiMI Base achieves a 20\% improvement over the Mistral Small 24B Base model on finance reasoning benchmark, while FiMI Instruct outperforms the Mistral Small 24B Instruct model by 87\% on domain-specific tool-calling. Moreover, FiMI achieves these significant domain gains while maintaining comparable performance to models of similar size on general benchmarks.
Cite
@article{arxiv.2602.05794,
title = {FiMI: A Domain-Specific Language Model for Indian Finance Ecosystem},
author = {Aboli Kathar and Aman Kumar and Anusha Kamath and Araveeti Srujan and Ashish Sharma and Chandra Bhushan and Divya Sorate and Duddu Prasanth Kumar and Evan Acharya and Harsh Sharma and Hrithik Kadam and Kanishk Singla and Keyur Doshi and Kiran Praveen and Kolisetty Krishna SK and Krishanu Adhikary and Lokesh MPT and Mayurdeep Sonowal and Nadeem Shaikh and Navya Prakash and Nimit Kothari and Nitin Kukreja and Prashant Devadiga and Rakesh Paul and Ratanjeet Pratap Chauhan and Raunak Kalani and Raviraj Joshi and Shamanth MH and Shantanu Pandey and Shubham Soni and Siddharth Dixit and Smriti Jopat and Sunil Patel and Suraj Singh and Suvradip Paul and Tulasi Pilla and Utkarsh Vaidya and Vineeth Nambiar and Vishal Kanvaty and Yatharth Dedhia},
journal= {arXiv preprint arXiv:2602.05794},
year = {2026}
}