FiMI: A Domain-Specific Language Model for Indian Finance Ecosystem

TL;DR

FiMI is a specialized Indian financial language model that enhances finance reasoning and tool-calling capabilities, tailored for Indian digital payment workflows, with significant improvements over baseline models.

Contribution

Developed FiMI, a domain-specific financial language model for India, with multi-stage training and fine-tuning for real-world financial workflows, outperforming general models in finance reasoning and tool use.

Findings

FiMI Base improves finance reasoning by 20% over baseline.

FiMI Instruct outperforms Mistral Instruct by 87% in tool-calling.

FiMI maintains performance on general benchmarks.

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.