Kaelum Bench

60-task benchmark

Accounting Knowledge

FMB-KNOW-ACC-01 · Public calibration

Foundational accounting concepts and financial statement mechanics.

What this benchmark tests

Published benchmark scope without claiming separate competency scores.

  • Balance sheet, income statement, equity, and cash flow
  • Journal entries and account logic
  • Accruals and subsequent measurement
  • Reconciliations and error diagnosis

Benchmark shape

Only definition and published scoring properties are shown.

Difficulty
12 basic · 30 intermediate · 18 advanced
Comparison
Configuration-specific
Available points
600 available points
AI judge
openai/gpt-5.6-sol
Deterministic component
487 points
AI rubric component
113 points

Published model runs

All 14 models

Select a row to inspect its benchmark-specific evidence.

  • Recorded cost & tokens
    Cost
    $1.15
    Tokens
    171,630
  • Recorded cost & tokens
    Cost
    $1.08
    Tokens
    146,307
  • Recorded cost & tokens
    Cost
    $0.9077
    Tokens
    208,196
  • Recorded cost & tokens
    Cost
    $1.72
    Tokens
    346,974
  • Recorded cost & tokens
    Cost
    $1.09
    Tokens
    309,508
  • Recorded cost & tokens
    Cost
    $2.16
    Tokens
    277,544
  • Recorded cost & tokens
    Cost
    $2.28
    Tokens
    249,444
  • Recorded cost & tokens
    Cost
    $1.89
    Tokens
    224,415
  • Recorded cost & tokens
    Cost
    $1.64
    Tokens
    185,914
  • Recorded cost & tokens
    Cost
    $1.63
    Tokens
    197,580
  • Recorded cost & tokens
    Cost
    $0.9641
    Tokens
    123,569
  • Recorded cost & tokens
    Cost
    $1.03
    Tokens
    131,370
  • Recorded cost & tokens
    Cost
    $1.41
    Tokens
    185,551
  • Recorded cost & tokens
    Cost
    $1.53
    Tokens
    182,452

Campaign evidence summary

Completed runs
14
Total recorded cost
$20.48
Total tokens
2,940,454
Valid outputs
836 / 840