RLTasksby ulam.ai
Models / Google

Google / MODEL EVIDENCE

Gemma 4 31B IT

16 benchmarks · 16 observations · 16 sources · released 2026-04-02

Scores at a glance

16 benchmarks · large dot: Gemma 4 31B IT; small dots: closest models where reported

GPQA Diamond ↗75.76%
9.2896.4

Gemma 4 31B IT · GPQA diamond · Best score (across scorers) · % ↗

CritPt ↗1.43%
032.29

Gemma 4 31B IT · CritPt · Accuracy · % ↗

LMCA ↗39.25%
2.7868.26

Gemma 4 31B IT · LMCA · Score · % ↗

DTBench ↗82.67%
41.6598.93

Gemma 4 31B IT · DTBench · Accuracy · % ↗

ALE-Bench ↗925.5 Performance points
137.782951.32

gemma-4-31b-it · Average contest performance ↗

ForecastBench ↗61.3 Brier index
46.467

gemma-4-31b-it† · Overall difficulty-adjusted Brier index ↗

Closest models by score

Similar positions on shared benchmark leaderboards. Smaller gaps are closer.

25.1 percentile-point gap

4 of 6 shared benchmark comparisons

27.1 percentile-point gap

5 of 6 shared benchmark comparisons

How similarity is measured

Mean absolute gap in percentile positions on shared benchmark settings; lower is closer. At least 60% overlap is required, with a minimum of three when available. Independent evaluations are used when there are at least three; otherwise published report comparisons are used. Repeated model observations within one setting are averaged before ranking. Writing-style traits and composite indices are excluded. This describes similarity, not an overall quality ranking.

Filter comparisons to this model →

Sources

Every plotted observation links to its evaluation. Reported configurations, benchmark versions and task subsets can differ; similarity describes shared scores, not training lift.

  1. Arena leaderboard — WebDev evaluation ↗

    Arena

  2. SciCode Benchmark | Artificial Analysis ↗

    Artificial Analysis

  3. DTBench — Conceptual Reasoning Index ↗

    Conceptual Reasoning Index

  4. LMCA — Conceptual Reasoning Index ↗

    Conceptual Reasoning Index

  5. Ale Bench — methodology and published results ↗

    Epoch AI · 2026

  6. CritPt v1 ↗

    Epoch AI

  7. Forecastbench — methodology and published results ↗

    Epoch AI · 2026

  8. GPQA Diamond ↗

    Epoch AI

  9. OTIS Mock AIME 2024-2025 ↗

    Epoch AI

  10. SimpleQA Verified ↗

    Epoch AI

  11. AI benchmarks & capabilities ↗

    Epoch AI

  12. WeirdML ↗

    Håvard Tveit Ihle

  13. CaseLaw — evaluation methodology and results ↗

    Vals AI · 2026

  14. MortgageTax — evaluation methodology and results ↗

    Vals AI · 2026

  15. SAGE — evaluation methodology and results ↗

    Vals AI · 2026

  16. Surface Evolver Bench ↗

    yhenon

Names used in source tables

Gemma 4 31B IT · gemma-4-31b-it · gemma-4-31b-it†