Medmarks - MMLU-Pro Health: leaderboard
The "MMLU-Pro Health" subtask of Medmarks. Living medical-model benchmark spanning verifiable and open-ended clinical evaluation families with model/system comparisons.
Metric: Score (%). Source: medmarks.ai. Saturation forecast: Estimated already saturated. 89 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Sonnet 4.5 | 83.86 | #142 |
| 2 | Gemini 3 Pro (Preview) | 83.74 | #66 |
| 3 | MiniMax-M2.1 | 82.97 | #289 |
| 4 | Grok 4 | 82.44 | #181 |
| 5 | GPT-5.1 (Medium) | 82.15 | #135 (GPT-5.1) |
| 6 | GPT-5.2 (Medium) | 81.38 | #110 (GPT-5.2) |
| 7 | GLM-4.7 FP8 | 80.07 | #209 |
| 8 | Qwen 3.6 35B A3B (Thinking) | 79.38 | #226 (Qwen 3.6 35B A3B) |
| 9 | Qwen 3.5 27B (Thinking) | 79.34 | #187 (Qwen 3.5 27B) |
| 10 | Gemma 4 31B (IT) | 79.14 | #176 |
| 11 | Qwen 3.5 35B A3B (Thinking) | 79.14 | #256 (Qwen 3.5 35B A3B) |
| 12 | Qwen 3 235B A22B (Thinking) | 78.69 | #313 (Qwen 3 235B A22B) |
| 13 | MiniMax-M2 | 78.44 | #317 |
| 14 | Qwen 3.6 27B (Thinking) | 78.28 | #170 (Qwen 3.6 27B) |
| 15 | Muse Glimmer 30B (xHigh) | 78 | #178 (Muse Glimmer 30B) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=medmarks-mmlu-pro-health · How It Works · Data refreshed daily, snapshot 2026-10-20.