Medmarks - MetaMedQA: leaderboard
The "MetaMedQA" 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: Around January 2027. 89 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.2 (Medium) | 82.88 | #110 (GPT-5.2) |
| 2 | GPT-5.1 (Medium) | 82.74 | #135 (GPT-5.1) |
| 3 | Gemini 3 Pro (Preview) | 81.67 | #66 |
| 4 | Grok 4 | 81.67 | #181 |
| 5 | Muse Glimmer 30B (xHigh) | 79.95 | #178 (Muse Glimmer 30B) |
| 6 | GLM-4.7 FP8 | 79.85 | #209 |
| 7 | Qwen 3.5 27B (Thinking) | 79.66 | #187 (Qwen 3.5 27B) |
| 8 | GPT-OSS-120B (High) | 79.39 | #341 (GPT-OSS-120B) |
| 9 | Qwen 3.6 27B (Thinking) | 79.24 | #170 (Qwen 3.6 27B) |
| 10 | Muse Glimmer 30B (High) | 79.15 | #178 (Muse Glimmer 30B) |
| 11 | Claude Sonnet 4.5 | 78.98 | #142 |
| 12 | Qwen 3.6 35B A3B (Thinking) | 78.66 | #226 (Qwen 3.6 35B A3B) |
| 13 | Qwen 3.5 35B A3B (Thinking) | 78.56 | #256 (Qwen 3.5 35B A3B) |
| 14 | Qwen 3 235B A22B (Thinking) | 78.03 | #313 (Qwen 3 235B A22B) |
| 15 | MiniMax-M2.1 | 77.32 | #289 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=medmarks-metamedqa · How It Works · Data refreshed daily, snapshot 2026-10-20.