Fin-R1 is a large language model(LLM) with a lightweight 7B parameter architecture that was created especially for the field of financial reasoning
Fin-R1 provides robust support for fundamental financial business scenarios in trusts, securities, banking, and insurance
Having built a data distillation framework based on DeepSeek-R1, closely adhering to the standard data processing parameter settings
Fin-R1 using Qwen2.5-7B-Instruct with SFT and reinforcement learning to increase accuracy and generalisation in financial reasoning tasks
Everyone employed a dual-reward system and the GRPO algorithm to maximise output format and accuracy after giving the model extensive reasoning abilities
It is just 3.0% behind DeepSeek-R1, and it is 6.0% better than the 70B-parameter DeepSeek-R1-Distill-Llama-70B (69.2)
A large language model for complicated financial reasoning, Fin-R1 was created and made publicly available by FinStep
DeepSeek-R1’s reasoning capabilities to financial scenarios and meet the needs for high-quality financial reasoning data