Med-CoReasoner - MMLU-ProX-Health - Japanese: leaderboard
Metric: Accuracy (%). Source: arxiv.org. Saturation forecast: Around December 2026. 9 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.2 (Low) | 73.94 |
| 2 | GPT-5.1 (Low) | 70.45 |
| 3 | GPT-4o | 67.39 |
| 4 | DeepSeek V3.2 | 66.81 |
| 5 | Claude 3.5 Haiku | 57.64 |
| 6 | Qwen 2.5 72B Instruct | 56.48 |
| 7 | Qwen 2.5 32B Instruct | 54.73 |
| 8 | Llama 3.1 70B Instruct | 48.33 |
Interactive version: theaggregate.ai/benchmark?slug=med-coreasoner-mmlu-prox-health-japanese · How It Works · Data refreshed daily, snapshot 2026-09-25.