The Fin AIOpen FinLLM Reasoning Leaderboard
Can reasoning-enhanced LLMs do finance? We test general-purpose and reasoning models on numerical question answering over reports, very long financial documents and XBRL equations. Accuracy in %, zero-shot.
What the numbers say
Who leads, task by task
Switch the metric to re-rank. Hover a bar for every score. Click a model type to hide it.
Where models struggle
Each dot is one model on one task. Hover or tap a dot to trace that model across all tasks.
Every score at a glance
All results
Built together
Acknowledgments
Datasets and tools this leaderboard builds on.
Method
Inference with the FinBen framework; answers scored with the DocMath-Eval evaluator. Average is the unweighted mean of the four tasks. Details in the Fino1 repository.
Add a model
Open an issue on The-FinAI/Fino1 with the model's Hub id and revision. It must be public, load with transformers AutoClasses and carry an open license.
Cite
@misc{qian2025fino1transferabilityreasoningenhancedllms,
title={Fino1: On the Transferability of Reasoning-Enhanced LLMs and Reinforcement Learning to Finance},
author={Lingfei Qian and Weipeng Zhou and Yan Wang and Xueqing Peng and Han Yi and Yilun Zhao and Jimin Huang and Qianqian Xie and Jian-yun Nie},
year={2025}, eprint={2502.08127}, archivePrefix={arXiv}, primaryClass={cs.CL},
url={https://arxiv.org/abs/2502.08127}
}