BenchmarksInternal Financial Modeling

Internal Financial Modeling

Completing financial modeling tasks in Moonshot’s internal finance evaluation.

FinanceRecipe not catalogued

Test definition

How the test is scored

The release chart labels K3.1 Pro at 57.6 and K3 at 50.4; task count, judge and aggregation are not specified.

Publisher-only internal result. It must not be merged with Frontier Finance or an independent finance-agent field.

A different task split, tool policy, judge or reasoning effort can change the result. Choose one comparison group below.

Select a publication and metric to compare the recorded field.

Comparison group

selfInternal finance and predecessor comparisons · Internal Financial Modeling

self% · Higher is better

Bars start at zero. 2 reported results in this comparison group; ties share a rank. Source ↗

Every numeric result, with its source and reported run conditions.

Find a resultSource

2 results

Results (%)
Model	Result (%)	Source	Run
Kimi K3.1 Pro	57.6	self	Details Labelled values from the internal finance chart. Task set, judge and aggregation are not published. Chart labels and explicit release prose, read 24 October 2026. These are internal publisher evaluations. The finance-chart Business Operation value differs from the full-table value and retains its own row. No Highspeed scores are inferred. Metric Internal Financial Modeling Published 23 October 2026 Effort Not reported separately Tasks / attempts Unknown / Unknown Confidence interval Not reported Build a reproduction plan →
Kimi K3	50.4	self	Details Labelled values from the internal finance chart. Task set, judge and aggregation are not published. Chart labels and explicit release prose, read 24 October 2026. These are internal publisher evaluations. The finance-chart Business Operation value differs from the full-table value and retains its own row. No Highspeed scores are inferred. Metric Internal Financial Modeling Published 23 October 2026 Effort Not reported separately Tasks / attempts Unknown / Unknown Confidence interval Not reported Build a reproduction plan →

Recipe not catalogued

Public execution instructions have not been verified in this catalogue. This status does not claim that the benchmark is closed or impossible to run.