OpenAI / MODEL EVIDENCE
1 benchmarks · 2 observations · 1 sources
1 benchmarks · large dot: GPT-4o (English subset); small dots: closest models where reported
GPT-4o (English subset) · WorldBench & CulturalBench · CulturalBench Easy · accuracy · full evaluation set ↗
62.4% · GPT-4o (English subset) · WorldBench & CulturalBench · CulturalBench Hard · accuracy · full evaluation set · CulturalBench: A Robust, Diverse, and Challenging Cultural Benchmark by Human-AI CulturalTeaming ↗
Similar positions on shared benchmark leaderboards. Smaller gaps are closer.
1 of 1 shared benchmark comparisons
1 of 1 shared benchmark comparisons
1 of 1 shared benchmark comparisons
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.
Every plotted observation links to its evaluation. Reported configurations, benchmark versions and task subsets can differ; similarity describes shared scores, not training lift.
Yu Ying Chiu, Liwei Jiang, Bill Yuchen Lin, Chan Young Park, Shuyue Stella Li, Sahithya Ravi, Mehar Bhatia, Maria Antoniak, Yulia Tsvetkov, Vered Shwartz, Yejin Choi
GPT-4o (English subset)