TheFinAI/Fino1-14B

Fino1-14B: Financial Reasoning Model

Fino1-14B, developed by TheFinAI, is a 14.8 billion parameter language model specifically fine-tuned for financial reasoning tasks. Built upon the Qwen2.5-14B-Instruct architecture, it leverages GPT-4o-generated financial reasoning paths from datasets like FinQA, TAT-QA, and DocMath-Eval, followed by reinforcement learning. This model is detailed in the paper "Fino1: On the Transferability of Reasoning-Enhanced LLMs and Reinforcement Learning to Finance" (arXiv:2502.08127).

Key Capabilities

• Specialized Financial Reasoning: Excels in numerical reasoning and question answering over financial documents, including text and tables.
• Enhanced Training: Utilizes advanced fine-tuning with GPT-4o-generated reasoning paths and reinforcement learning for improved accuracy in financial contexts.
• High Context Length: Features a substantial context length of 131,072 tokens, enabling processing of long financial documents.
• Research Baseline: Serves as a strong baseline for further research and development in financial AI models.

Good For

• Financial Numerical Reasoning Research: Ideal for academic and industrial research into AI applications for finance.
• Question Answering on Financial Data: Performing question answering over complex financial texts and tables.
• Benchmarking Financial LLMs: Evaluating and comparing new financial reasoning models against a specialized baseline.

It's important to note that while powerful, Fino1-14B is intended for research and should not be used for investment advice or automated financial decisions without human oversight, as it may produce incorrect calculations or hallucinations.