# FinBen

A broad set of financial NLP tasks across seven areas: information extraction, textual analysis, QA, text generation, risk management, forecasting and decision-making. It includes a stock-trading simulation.

## Facts

- Maker: The Fin AI (Qianqian Xie, Jimin Huang et al., 34 authors)
- Year: 2024
- Use case: Finance knowledge
- Size: 36 datasets, 24 tasks
- Public split: Datasets, code and results through the PIXIU repo and Hugging Face. Each dataset keeps its own licence.
- Grading: Per-task automatic metrics: F1, accuracy, MCC, entity F1, exact match, ROUGE, BERTScore and BARTScore. Trading uses cumulative return, Sharpe ratio and drawdown. There is no aggregate score.
- Licence: MIT (code); data licences vary per dataset (MIT, CC BY 4.0, CC BY-NC 4.0, CC BY-SA 3.0, CC BY-NC-SA 4.0)
- Submission: Not stated
- Products allowed: No
- Latest version: arXiv v2 (2024-06-19)
- What it does not cover: No single headline score, no tool-using agents, and no live market data retrieval.
- Last checked: 1 Nov 2026
- Links: paper https://arxiv.org/abs/2402.12659, code https://github.com/The-FinAI/PIXIU, leaderboard https://huggingface.co/spaces/TheFinAI/Open-FinLLM-Leaderboard
- Directory page: https://benchmarks.finance/benchmarks/finben/
- Leaderboard: https://benchmarks.finance/leaderboards/finben/

## Results

No scores recorded here yet.

Source: https://benchmarks.finance/benchmarks/finben/

Created and maintained by Moneda.