Moneda Bench, made by Moneda
Moneda KnowledgeBench
Whether a model knows finance: valuation, fixed income, accounting, derivatives, portfolio theory and market mechanics, plus whether it can explain a concept to a beginner.
First results
Coming
About this benchmark
- Size
- 400 items: 300 multiple choice and numeric, 100 beginner explanations
- Public split
- Yes, about a third held back, plus a randomised-answer twin to detect training on the public file
- Grading
- Code for multiple choice and numeric answers. A rubric for explanations (terms defined, numbers given context) with a readability score.
- Licence
- Items CC BY 4.0, harness MIT
- Submission
- Open harness at release
- Products allowed
- No
- What it does not cover
- Tool use, live data and judgement about a specific person's situation. Models only, no tools.
How it will run
Every item is written fresh, not copied from exam sets or textbooks, and reviewed by a finance professional. An answer key is kept only when models from two different labs and the reviewer agree. The public file ships with a twin whose keys are randomised, so a model scoring well above the twin's ceiling has likely seen the file in training.
Status: first results coming.
Results
First results coming.
Papers and links
Paper, data and harness will be linked here at release.
Last checked 1 Nov 2026. Something out of date or wrong? Send a correction.