FinQA Dataset
by Zhiyu Chen et al. / University of California Santa Barbara · open-source · Last verified 2026-03-17
FinQA is a dataset for numerical reasoning over financial data, containing 8,281 financial question-answer pairs derived from earnings reports of S&P 500 companies. Each example requires multi-step numerical reasoning over a combination of unstructured text and structured tables, making it one of the most challenging benchmarks for financial AI systems.
https://github.com/czyssrs/FinQA ↗B
B—Above Average
Adoption: BQuality: A+Freshness: B+Citations: B+Engagement: F
Specifications
- License
- MIT
- Pricing
- open-source
- Capabilities
- financial-question-answering, numerical-reasoning, table-text-reasoning
- Integrations
- HuggingFace Datasets
- Use Cases
- financial-nlp-research, benchmark, model-training
- API Available
- No
- Tags
- financial-qa, numerical-reasoning, table-qa, earnings-reports, benchmark
- Added
- 2026-03-17
- Completeness
- 100%
Index Score
65.1Adoption
68
Quality
92
Freshness
72
Citations
78
Engagement
0
Put AI to work for your business
Deploy this dataset alongside autonomous AaaS agents that handle tasks end-to-end — no babysitting required.