TreasuryBench - A public benchmark for AI financial advice

by
We tested Treasury, Origin, and Monarch on the same real personal-finance questions, scored by an independent AI judge. Treasury scored 89 to Origin’s 71 and Monarch’s 52 — winning all 12 categories with zero dangerous answers. The methodology, scoring code, and raw results are open source.

Add a comment

Replies

Best
Maker
📌
AI money advice is only useful if it is grounded in your actual financial life — and safe to act on. We built TreasuryBench to make that test visible: the same connected accounts, the same questions, and a published record of every answer. Treasury’s August run scored 89/100 with zero dangerous answers. Origin and Monarch’s comparison runs were measured in June, so they may have improved since; the raw data is public so anyone can audit the result or open an issue.