Benchmark menu

Model results

Kimi K3

See how Kimi K3 performed, which reasoning settings were tested, how often it produced a working player program, and what those runs cost.

Kimi K3

Provider: Moonshot

Rank
10
Status
Partial 3/3
Rank change
Baseline
Score
50.0
Last tested
2026-08-22
Settings tested
All settings tested

Why this result is partial: the official generation cohort is incomplete. The rank uses the available reasoning-setting results.

Scores by reasoning setting

Available settings

LowHighMax

Checked on 2026-07-24 · Provider documentation

Scores from tested settings

Tested settingScoreResult statusDisplay group
Low53.4Tested and includedBaseline
High47.6Tested and includedBalanced
Max48.9Tested and includedIntensive

How settings are grouped

  • Baseline: Low — tested and included
  • Balanced: High — tested and included
  • Intensive: Max — tested and included

Player program results

Max: 14% / 57% / 29% · n 7High: 88% / 12% / 0% · n 8Low: 88% / 12% / 0% · n 8 All settings testedMany programs needed repairMany programs failed

Generation cost

Average estimated cost
$0.30
Median estimated cost
$0.30
Estimated range
$0.06–$0.63
Cost data
24 combinations / 24 recorded runs
Observed total
$7.14
Price source
Provider-reported
Price list
boardgame-list-prices-2026-08-08
Average output
18k tokens
Output range
2.2k–40k tokens
Samples
24

Estimated cost gives equal weight to each model, game, and reasoning-setting combination.

Test setup

  • Combines this model's tested reasoning settings
  • Uses the standard GameBench player-program prompt