CCBench-Health (Explicit Norms): leaderboard

Metric: Cultural Competence Score (%; harmonic mean of the follow rate and the avoid rate, the shares of checklist recommendations satisfied for norms a persona follows or avoids; 60 personas with graded norm-adherence states across six cultures (Afghan, Burmese, Chinese, Maori, Nepali, Vietnamese), 52 real health questions each (3,120 interactions); GPT-5.2 checklist evaluator; temperature 0.7; upper-bound setting with the persona norm-adherence states stated explicitly in the system prompt)). Source: arxiv.org. Saturation forecast: Around August 2028. 5 models tracked.

Top models

#ModelScore
1Gemini 2.5 Pro31.8
2GPT-5.230.9
3Llama 3.2 90B18.5

Interactive version: theaggregate.ai/benchmark?slug=ccbench-health-explicit-norms · How It Works · Data refreshed daily, snapshot 2026-09-29.