Big Finance Bench: Results as Published

BigFinanceBench tests whether an AI agent can do the research work of a professional financial analyst. Each of its 928 questions comes with a reference answer and a rubric that breaks the derivation into steps that can be checked one by one, so partial credit follows the analyst workflow rather than the final number alone. A 50-question subset is public; the full set stays private.

Every number here is read from the same catalog file as the benchmark table, data version 2026103001, and is the number the table shows for that model and test. Where one page printed a model under several setups, each value is listed. The catalog is public in armanobosyan/sugra-llm-benchmarks; a wrong number can be reported from the table, with Report an error under the value, or as an issue in that repository.