RLTasksby ulam.ai
Models / Google

Google / MODEL EVIDENCE

Gemini 3.1 Flash-Lite

17 benchmarks · 24 observations · 17 sources · released 2026-03-03

Scores at a glance

17 benchmarks · large dot: Gemini 3.1 Flash-Lite; small dots: closest models where reported

IFBench ↗77.21%
75.5883.33

Gemini 3.1 Flash-Lite · IFBench · AA implementation ↗

CritPt ↗1.14%
032.29

Gemini 3.1 Flash-Lite · CritPt · Accuracy · % ↗

LMCA ↗35.01%
2.7868.26

Gemini 3.1 Flash-Lite · LMCA · Score · % ↗

DTBench ↗76.8%
41.6598.93

Gemini 3.1 Flash-Lite · DTBench · Accuracy · % ↗

SciCode ↗43.4%
1.2766.9

Gemini 3.1 Flash-Lite · SciCode · Score · % ↗

MCP Atlas ↗57.10%
40.288.1

gemini-3.1-flash-lite (high) · Scale Labs · Score ↗

ForecastBench ↗62.7 Brier index
46.467

gemini-3.1-flash-lite† · Overall difficulty-adjusted Brier index ↗

Closest models by score

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

49.6 percentile-point gap

3 of 3 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. APEX-Agents-AA Benchmark | Artificial Analysis ↗

    Artificial Analysis

  3. IFBench Benchmark | Artificial Analysis ↗

    Artificial Analysis

  4. SciCode Benchmark | Artificial Analysis ↗

    Artificial Analysis

  5. DTBench — Conceptual Reasoning Index ↗

    Conceptual Reasoning Index

  6. LMCA — Conceptual Reasoning Index ↗

    Conceptual Reasoning Index

  7. CritPt v1 ↗

    Epoch AI

  8. Forecastbench — methodology and published results ↗

    Epoch AI · 2026

  9. FrontierMath Tiers 1-3 (v2) ↗

    Epoch AI

  10. GPQA Diamond ↗

    Epoch AI

  11. Humanity's Last Exam ↗

    Epoch AI

  12. OTIS Mock AIME 2024-2025 ↗

    Epoch AI

  13. WeirdML (v2) ↗

    Epoch AI

  14. AI benchmarks & capabilities ↗

    Epoch AI

  15. FutureSearch Evals ↗

    FutureSearch

  16. EnigmaEval ↗

    Scale Labs / SEAL

  17. Scale Labs — evaluation methodology and results ↗

    Scale Labs / SEAL · 2026

Names used in source tables

Gemini 3.1 Flash-Lite · gemini-3.1-flash-lite (high) · gemini-3.1-flash-lite†