RLTasksby ulam.ai
Models / OpenAI

OpenAI / MODEL EVIDENCE

o3-2025-04-16

2 benchmarks · 4 observations · 2 sources

Scores at a glance

2 benchmarks · large dot: o3-2025-04-16; small dots: closest models where reported

FrontierMath-2025-02-28-Public ↗10%
070

o3-2025-04-16_high · FrontierMath-2025-02-28-Public · FrontierMath public · 10 questions · source run score · Epoch AI · public February 2025 sample · historical release ↗

2 more reported results

10% · o3-2025-04-16_low · FrontierMath-2025-02-28-Public · FrontierMath public · 10 questions · source run score · Epoch AI · public February 2025 sample · historical release · A Rosetta Stone for AI Benchmarks — released Epoch evaluation runs ↗

0% · o3-2025-04-16_medium · FrontierMath-2025-02-28-Public · FrontierMath public · 10 questions · source run score · Epoch AI · public February 2025 sample · historical release · A Rosetta Stone for AI Benchmarks — released Epoch evaluation runs ↗

ForecastBench ↗64.6 Brier index
46.467

o3-2025-04-16† · Overall difficulty-adjusted Brier index ↗

Closest models by score

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

There are not enough shared scored settings to identify close comparisons reliably.

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. Forecastbench — methodology and published results ↗

    Epoch AI · 2026

  2. A Rosetta Stone for AI Benchmarks — released Epoch evaluation runs ↗

    Epoch AI · 2025

Names used in source tables

o3-2025-04-16_high · o3-2025-04-16_low · o3-2025-04-16_medium · o3-2025-04-16†