RLTasksby ulam.ai
Models / xAI

xAI / MODEL EVIDENCE

Grok 2 (Dec 2024)

10 benchmarks · 10 observations · 10 sources · released 2024-12-12

Scores at a glance

10 benchmarks · large dot: Grok 2 (Dec 2024); small dots: closest models where reported

GPQA Diamond ↗53.79%
9.2896.4

Grok-2 (Dec 2024) · GPQA diamond · Best score (across scorers) · % ↗

MATH Level 5 ↗63.52%
3.2998.13

Grok-2 (Dec 2024) · MATH level 5 · Best score (across scorers) · % ↗

DTBench ↗65.21%
41.6598.93

Grok-2 (Dec 2024) · DTBench · Accuracy · % ↗

LiveBench ↗54.3%
22.0882.35

Grok-2 (Dec 2024) · LiveBench · Global average · % ↗

Closest models by score

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

Phi-4 →Microsoft
1.8 percentile-point gap

3 of 4 shared benchmark comparisons

2.2 percentile-point gap

3 of 4 shared benchmark comparisons

3.2 percentile-point gap

4 of 4 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. DTBench — Conceptual Reasoning Index ↗

    Conceptual Reasoning Index

  2. FrontierMath Tiers 1-3 (v1, old) ↗

    Epoch AI

  3. GPQA Diamond ↗

    Epoch AI

  4. MATH Level 5 ↗

    Epoch AI

  5. OTIS Mock AIME 2024-2025 ↗

    Epoch AI

  6. SimpleBench ↗

    Epoch AI

  7. AI benchmarks & capabilities ↗

    Epoch AI

  8. Writing benchmark — incorporating mandatory story elements ↗

    lechmazur

  9. WeirdML ↗

    Håvard Tveit Ihle

  10. LiveBench — model evaluation leaderboard ↗

    LiveBench Team

Names used in source tables

Grok-2 (Dec 2024)