The Fin AIOpen Financial LLM Leaderboard
Real financial work mixes languages and media: filings, scanned statements, earnings calls. MultiFinBen tests LLMs on 36 datasets in five languages and three modalities — text, vision and audio.
What the numbers say
Who leads, and in which modality
The modality-balanced score is the mean of a model's text, vision and audio averages, so a model that cannot see or hear scores 0 there. Switch to a single modality to compare only the models that support it.
What's inside MultiFinBen
Every tile is a dataset; its fill shows the best score any model reached. Hover for the top three, click to open it below.
Same model, different language
Average text score per language for every model that handles text. BI is the bilingual DOLFIN task; MU are the multilingual PolyFiQA tasks, which mix sources in several languages.
Pick a dataset
All results
Standardized scores (0–100) from MultiFinBen Table 3. A 0 can mean the model scored nothing or does not support that input.
Built together
Author affiliations on the two papers behind this page.
Leaderboard · Open FinLLM Leaderboard: Towards Financial AI Readiness
Benchmark · MultiFinBen
Method
Datasets are chosen with a difficulty-aware selection that keeps one dataset per modality, language and task tier. Evaluation code is in the MultiFinBen repository; scores are standardized to 0–100.
Add a model
Open an issue on MultiFinBen with the model's Hub id and revision. We run the full benchmark and add the scores here.
Cite
@misc{lin2025openfinllmleaderboardfinancial,
title={Open FinLLM Leaderboard: Towards Financial AI Readiness},
author={Shengyuan Colin Lin and Felix Tian and Keyi Wang and Xingjian Zhao and Jimin Huang and Qianqian Xie and Luca Borella and Matt White and Christina Dan Wang and Kairong Xiao and Xiao-Yang Liu Yanglet and Li Deng},
year={2025}, eprint={2501.10963}, archivePrefix={arXiv}, primaryClass={cs.CE},
url={https://arxiv.org/abs/2501.10963}
}
@misc{peng2025multifinbenbenchmarkinglargelanguage,
title={MultiFinBen: Benchmarking Large Language Models for Multilingual and Multimodal Financial Application},
author={Xueqing Peng and Lingfei Qian and Yan Wang and Ruoyu Xiang and Yueru He and Yang Ren and Mingyang Jiang and Vincent Jim Zhang and Yuqing Guo and Jeff Zhao and Huan He and Yi Han and Yun Feng and Yuechen Jiang and Yupeng Cao and Haohang Li and Yangyang Yu and Xiaoyu Wang and Penglei Gao and Shengyuan Lin and Keyi Wang and Shanshan Yang and Yilun Zhao and Zhiwei Liu and Peng Lu and Jerry Huang and Suyuchen Wang and Triantafillos Papadopoulos and Polydoros Giannouris and Efstathia Soufleri and Nuo Chen and Zhiyang Deng and Heming Fu and Yijia Zhao and Mingquan Lin and Meikang Qiu and Kaleb E Smith and Arman Cohan and Xiao-Yang Liu and Jimin Huang and Guojun Xiong and Alejandro Lopez-Lira and Xi Chen and Junichi Tsujii and Jian-Yun Nie and Sophia Ananiadou and Qianqian Xie},
year={2025},
eprint={2506.14028},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2506.14028},
}