RLTasksby ulam.ai
Models / Google

Google / MODEL EVIDENCE

Gemma 3 27B

14 benchmarks · 17 observations · 14 sources

Scores at a glance

14 benchmarks · large dot: Gemma 3 27B; small dots: closest models where reported

GPQA Diamond ↗47.73%
9.2896.4

Gemma 3 27B · GPQA diamond · Best score (across scorers) · % ↗

MATH Level 5 ↗74.04%
3.2998.13

Gemma 3 27B · MATH level 5 · Best score (across scorers) · % ↗

LMCA ↗12.28%
2.7868.26

Gemma 3 27B · LMCA · Score · % ↗

DTBench ↗52.53%
41.6598.93

Gemma 3 27B · DTBench · Accuracy · % ↗

LiveBench ↗49.99%
22.0882.35

Gemma 3 27B · LiveBench · Global average · % ↗

LiveCodeBench Pro ↗601 Elo
1999

Gemma 3 27B · LiveCodeBench Pro · Elo rating · human contest calibration ↗

3 more reported results

0% · Gemma 3 27B · LiveCodeBench Pro · Hard · pass@1 · difficulty breakdown · LiveCodeBench Pro: How Do Olympiad Medalists Judge LLMs in Competitive Programming? ↗

0% · Gemma 3 27B · LiveCodeBench Pro · Medium · pass@1 · difficulty breakdown · LiveCodeBench Pro: How Do Olympiad Medalists Judge LLMs in Competitive Programming? ↗

8.5% · Gemma 3 27B · LiveCodeBench Pro · Easy · pass@1 · difficulty breakdown · LiveCodeBench Pro: How Do Olympiad Medalists Judge LLMs in Competitive Programming? ↗

Closest models by score

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

2.7 percentile-point gap

3 of 3 shared benchmark comparisons

4.4 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. Aider LLM Leaderboards — Polyglot coding benchmark ↗

    Aider

  2. SciCode Benchmark | Artificial Analysis ↗

    Artificial Analysis

  3. LiveCodeBench Pro: How Do Olympiad Medalists Judge LLMs in Competitive Programming? — published result table ↗

    Zihan Zheng, Zerui Cheng, Zeyu Shen, Shang Zhou, Kaiyuan Liu, Hansen He et al. · 2025

    All 19 authors

    Zihan Zheng, Zerui Cheng, Zeyu Shen, Shang Zhou, Kaiyuan Liu, Hansen He, Dongruixuan Li, Stanley Wei, Hangyi Hao, Jianzhu Yao, Peiyao Sheng, Zixuan Wang, Wenhao Chai, Aleksandra Korolova, Peter Henderson, Sanjeev Arora, Pramod Viswanath, Jingbo Shang, Saining Xie

  4. DTBench — Conceptual Reasoning Index ↗

    Conceptual Reasoning Index

  5. LMCA — Conceptual Reasoning Index ↗

    Conceptual Reasoning Index

  6. CritPt v1 ↗

    Epoch AI

  7. GPQA Diamond ↗

    Epoch AI

  8. MATH Level 5 ↗

    Epoch AI

  9. OTIS Mock AIME 2024-2025 ↗

    Epoch AI

  10. AI benchmarks & capabilities ↗

    Epoch AI

  11. Fiction.liveBench — March 14, 2025 benchmark release ↗

    Fiction.live · 2025

  12. GeoBench — GeoGuessr model benchmark ↗

    GeoBench

  13. Writing benchmark — incorporating mandatory story elements ↗

    lechmazur

  14. LiveBench — model evaluation leaderboard ↗

    LiveBench Team

Release dates reported by sources

2025-03-11 · 2025-03-12

Names used in source tables

Gemma 3 27B