CCBench-Health - Avoid Rate: leaderboard

Metric: Avoid rate (%; share of checklist recommendations satisfied for norms the persona avoids, a measure of stereotype resistance; 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; the persona conversation history is given before the question and the norms must be inferred from it)). Source: arxiv.org. Saturation forecast: Around September 2028. 5 models tracked.

Top models

#ModelScore
1GPT-5.251.7
2Gemini 2.5 Pro45.3
3Qwen 3.5 397B A17B43.7
4Llama 3.2 90B37.9

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