RLTasksby ulam.ai
Models / Google

Google / MODEL EVIDENCE

Gemini 2.5 Pro Exp (Mar 2025)

10 benchmarks · 10 observations · 10 sources · released 2025-03-25

Scores at a glance

10 benchmarks · large dot: Gemini 2.5 Pro Exp (Mar 2025); small dots: closest models where reported

GPQA Diamond ↗83.84%
9.2896.4

Gemini 2.5 Pro Exp (Mar 2025) · GPQA diamond · Best score (across scorers) · % ↗

BALROG ↗43.3%
3.768.3

Gemini 2.5 Pro Exp (Mar 2025) · Balrog · Average progress · % ↗

GeoBench ↗81%
3286

Gemini 2.5 Pro Exp (Mar 2025) · GeoBench · ACW Country % · % ↗

SimpleBench ↗51.6%
10.781.9

Gemini 2.5 Pro Exp (Mar 2025) · SimpleBench · Score (AVG@5) · % ↗

LiveBench ↗82.35%
22.0882.35

Gemini 2.5 Pro Exp (Mar 2025) · LiveBench · Global average · % ↗

EnigmaEval ↗4.14%
0.3839.28

Gemini 2.5 Pro Exp (Mar 2025) · EnigmaEval · Accuracy · % ↗

Closest models by score

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

16.4 percentile-point gap

7 of 10 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. BALROG: Benchmarking Agentic LLM and VLM Reasoning On Games ↗

    Paglieri, Davide, Cupiał, Bartłomiej, Coward, Samuel, Piterbarg, Ulyana, Wolczyk, Maciej, Khan, Akbir et al. · 2024

    All 13 authors

    Paglieri, Davide, Cupiał, Bartłomiej, Coward, Samuel, Piterbarg, Ulyana, Wolczyk, Maciej, Khan, Akbir, Pignatelli, Eduardo, Kuciński, Łukasz, Pinto, Lerrel, Fergus, Rob, Foerster, Jakob Nicolaus, Parker-Holder, Jack, Rocktäschel, Tim

  3. GPQA Diamond ↗

    Epoch AI

  4. Humanity's Last Exam ↗

    Epoch AI

  5. SimpleBench ↗

    Epoch AI

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

    Fiction.live · 2025

  7. GeoBench — GeoGuessr model benchmark ↗

    GeoBench

  8. Writing benchmark — incorporating mandatory story elements ↗

    lechmazur

  9. EnigmaEval ↗

    Scale Labs / SEAL

  10. LiveBench — model evaluation leaderboard ↗

    LiveBench Team

Names used in source tables

Gemini 2.5 Pro Exp (Mar 2025)