RLTasksby ulam.ai
Models / OpenAI

OpenAI / MODEL EVIDENCE

GPT-5 Pro

14 benchmarks · 16 observations · 15 sources · released 2025-10-07

Scores at a glance

14 benchmarks · large dot: GPT-5 Pro; small dots: closest models where reported

Closest models by score

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

8.6 percentile-point gap

4 of 6 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. Algotune — methodology and published results ↗

    Epoch AI · 2026

  3. ARC-AGI-1 ↗

    Epoch AI

  4. ARC-AGI-2 ↗

    Epoch AI

  5. Chess Puzzles ↗

    Epoch AI

  6. FrontierMath Tier 4 (v1, old) ↗

    Epoch AI

  7. FrontierMath Tier 4 (v2) ↗

    Epoch AI

  8. FrontierMath Tiers 1-3 (v2) ↗

    Epoch AI

  9. Humanity's Last Exam ↗

    Epoch AI

  10. SimpleBench ↗

    Epoch AI

  11. AI benchmarks & capabilities ↗

    Epoch AI

  12. WeirdML ↗

    Håvard Tveit Ihle

  13. EnigmaEval ↗

    Scale Labs / SEAL

  14. Scale Labs — evaluation methodology and results ↗

    Scale Labs / SEAL · 2026

  15. Scale Labs — evaluation methodology and results ↗

    Scale Labs / SEAL · 2026

Names used in source tables

GPT-5 Pro · GPT-5 Pro (medium) · gpt-5-pro