BigFinanceBench: A Workflow-Grounded Benchmark for Financial-Research Agents
For financial-research practitioners and AI developers, this benchmark provides a more comprehensive evaluation of agent workflows, highlighting that current models fall short in auditable derivation quality.
BigFinanceBench introduces a 928-item benchmark for financial-research agents that evaluates the full derivation of answers via point-weighted rubrics, not just final outputs. The best system achieves only 58.8% rubric score, revealing substantial headroom.
Financial-research answers are decision-relevant only when another analyst can audit how they were produced: which source was chosen, which period and accounting definition were used, which assumptions were made, and how the calculation was performed. Existing finance benchmarks largely evaluate isolated subskills or final answers, leaving the auditable derivation itself under-measured. We introduce BigFinanceBench, a 928-item expert-authored benchmark of open-ended financial-research tasks in which each item pairs a ground-truth reference answer with a point-weighted rubric that decomposes the derivation into independently checkable steps. BigFinanceBench is workflow-grounded in that it evaluates the full derivation rather than only the final output. Across 36,241 rubric points, the benchmark supports partial-credit evaluation and localization of failures across the analyst workflow. Evaluating ten current frontier and open-weight agents, we find substantial headroom: the best system reaches only 58.8% rubric score, final-answer accuracy is a useful but lossy proxy for derivation quality, and model capability varies non-uniformly across financial workflows.