RLTasksby ulam.ai
Models / Google

Google / MODEL EVIDENCE

Gemini 3 Deep Think

3 benchmarks · 6 observations · 4 sources · released 2026-02-12

Scores at a glance

3 benchmarks · large dot: Gemini 3 Deep Think; small dots: closest models where reported

CritPt ↗25.71%
032.29

Gemini 3 Deep Think · CritPt · Accuracy · % ↗

Closest models by score

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

2.8 percentile-point gap

3 of 3 shared benchmark comparisons

4.6 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. ARC Prize - Leaderboard ↗

    ARC Prize Foundation

  2. ARC-AGI-1 ↗

    Epoch AI

  3. ARC-AGI-2 ↗

    Epoch AI

  4. CritPt v1 ↗

    Epoch AI

Names used in source tables

Gemini 3 Deep Think