Medmarks - LongHealth Task 2: leaderboard
The "LongHealth Task 2" 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 | Qwen 3.6 27B (Thinking) | 91.33 | #170 (Qwen 3.6 27B) |
| 2 | Gemini 3 Pro (Preview) | 90.96 | #66 |
| 3 | Qwen 3.5 35B A3B (Thinking) | 90.75 | #256 (Qwen 3.5 35B A3B) |
| 4 | Qwen 3.6 35B A3B (Thinking) | 90.75 | #226 (Qwen 3.6 35B A3B) |
| 5 | Claude Sonnet 4.5 | 90.58 | #142 |
| 6 | Qwen 3 Next 80B A3B Instruct | 90.33 | #368 |
| 7 | GPT-5.1 (Medium) | 90.33 | #135 (GPT-5.1) |
| 8 | Qwen 3 235B A22B (Thinking) | 90.25 | #313 (Qwen 3 235B A22B) |
| 9 | Qwen 3.5 27B (Thinking) | 90.17 | #187 (Qwen 3.5 27B) |
| 10 | Gemma 4 31B (IT) | 89.54 | #176 |
| 11 | Ministral-3-8B-Instruct-2512 | 89.29 | #661 |
| 12 | Magistral Small | 88.96 | #1016 |
| 13 | Nemotron 3.5 Lightning | 88.92 | #467 |
| 14 | Ministral-3-14B-Instruct-2512 | 88.88 | #615 |
| 15 | GLM-4.7 FP8 | 88.75 | #209 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=medmarks-longhealth-task-2 · How It Works · Data refreshed daily, snapshot 2026-10-20.