Coverage for src / quber / playground / benchmark / __init__.py: 100%

1"""FinanceBench benchmark: gold questions, answers, and evidence in Postgres. 

2 

3The benchmark drives the playground corpus: every ingested document takes its 

4identity from a benchmark `doc_name` (see `load.doc_key_for`), and evaluation 

5reads gold answers from the `financebench` schema — which retrieval and the 

6answer agents never touch. 

7"""