Benchmark menu

Model results

Gemma 4 31B

See how Gemma 4 31B performed, which reasoning settings were tested, how often it produced a working player program, and what those runs cost.

Gemma 4 31B

Provider: Google

Rank
31
Status
Partial 2/3
Rank change
Baseline
Score
34.6
Last tested
2026-08-05
Settings tested
Intensive not supported

Why this result is partial: one or more bracket scores are missing; three distinct provider settings are not documented; one or more documented settings are untested or ineligible; generation-quality evidence is missing; the official generation cohort is incomplete. The rank uses the available reasoning-setting results.

Scores by reasoning setting

Available settings

Reasoning disabledReasoning enabled

Checked on 2026-07-30 · Provider documentation

Scores from tested settings

Tested settingScoreResult statusDisplay group
Reasoning disabled31.1Tested and includedBaseline
Reasoning enabled38.0Tested and includedBalanced

How settings are grouped

  • Baseline: Reasoning disabled — tested and included
  • Balanced: Reasoning enabled — tested and included
  • Intensive: Not available for this model

Player program results

Reasoning enabled: 0% / 57% / 43% · n 7Reasoning disabled: 72% / 14% / 14% · n 7 Intensive not supportedMany programs needed repairMany programs failed

Generation cost

Average estimated cost
$0.0045
Median estimated cost
$0.0042
Estimated range
$0.0014–$0.0092
Cost data
14 combinations / 14 recorded runs
Observed total
$0.06
Price source
Provider-reported
Price list
Not recorded for this older data
Average output
5.5k tokens
Output range
1.3k–13k tokens
Samples
14

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