# Numa 1 finance AI benchmark results

Every finance benchmark score for Numa 1 by Moneda: 6 benchmarks, each with its rank, metric and source date. Scores are never combined across benchmarks.

## About

Numa 1 is the model behind Moneda AI, the agent in the Moneda app. It researches stocks, ETFs and companies with live market data and reads the filings itself. Scores here are from runs Moneda did on the product as users get it.

https://moneda.com/models/numa-1

## Board standings

Numa 1 is first on 1 benchmark, top 3 on 2 benchmarks, scored on 6 benchmarks. Places on each benchmark's main board with 2 or more models; scores are never combined. https://benchmarks.finance/method/#standings
- FinanceBench: #1 of 5
- FinSearchComp: #3 of 11

## Scores

| Benchmark | Rank | Score | Metric | Source | Date |
| --- | --- | --- | --- | --- | --- |
| [FinanceBench](https://benchmarks.finance/benchmarks/financebench/) | 1 of 5 | 67 | % correct, no filing given (150-question sample) | Moneda run, 2026-11-02 | 2 Nov 2026 |
| [FinSearchComp](https://benchmarks.finance/benchmarks/finsearchcomp/) | 3 of 11 | 59.4 | Global subset average accuracy % | Moneda run, 2026-11-01 | 1 Nov 2026 |
| [FinanceQA](https://benchmarks.finance/benchmarks/afterquery-financeqa/) | 4 of 5 | 0.355 | accuracy (total, 0-1) | Moneda run, 2026-11-01 | 1 Nov 2026 |
| [Finance Agent Benchmark](https://benchmarks.finance/benchmarks/vals-finance-agent/) | 1 of 1 | 80.2 | Partial credit % on the 27 public questions (run by Moneda; not comparable, unranked) | Moneda run, 2026-10-09 | 9 Oct 2026 |
| [Moneda FilingsBench](https://benchmarks.finance/benchmarks/moneda-filingsbench/) | 1 of 1 | 91.6 | % correct (private split) | Moneda run, 2026-11-02 | 2 Nov 2026 |
| [Moneda PortfolioBench](https://benchmarks.finance/benchmarks/moneda-portfoliobench/) | 1 of 1 | 96 | Portfolio tasks, % correct | Moneda run, 2026-11-01 | 1 Nov 2026 |

Source: https://benchmarks.finance/models/numa-1/

Created and maintained by Moneda.