Med-CoReasoner - MMLU-ProX-Health - Chinese: leaderboard

Metric: Accuracy (%). Source: arxiv.org. Saturation forecast: Around December 2026. 9 models tracked.

Top models

#ModelScore
1GPT-5.2 (Low)72.63
2GPT-5.1 (Low)71.32
3GPT-4o67.39
4DeepSeek V3.266.81
5Qwen 2.5 72B Instruct59.39
6Qwen 2.5 32B Instruct58.22
7Claude 3.5 Haiku56.33
8Llama 3.1 70B Instruct51.09

Interactive version: theaggregate.ai/benchmark?slug=med-coreasoner-mmlu-prox-health-chinese · How It Works · Data refreshed daily, snapshot 2026-09-25.