Medmarks - SCTpublic: leaderboard
The "SCTpublic" 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-5.1 (Medium) | 77.96 | #135 (GPT-5.1) |
| 2 | GPT-5.2 (Medium) | 74.26 | #110 (GPT-5.2) |
| 3 | Gemma 4 31B (IT) | 74.08 | #176 |
| 4 | Grok 4 | 73.57 | #181 |
| 5 | Qwen 3.6 35B A3B (Thinking) | 73.45 | #226 (Qwen 3.6 35B A3B) |
| 6 | Claude Sonnet 4.5 | 72.29 | #142 |
| 7 | Nemotron 3.5 Lightning | 71.86 | #467 |
| 8 | Qwen 3.5 27B (Thinking) | 71.61 | #187 (Qwen 3.5 27B) |
| 9 | GPT-OSS-120B (Medium) | 71.35 | #341 (GPT-OSS-120B) |
| 10 | Muse Glimmer 30B (xHigh) | 70.92 | #178 (Muse Glimmer 30B) |
| 11 | GPT-OSS-120B (Low) | 70.41 | #341 (GPT-OSS-120B) |
| 12 | MiniMax-M2.1 | 70.01 | #289 |
| 13 | MiniMax-M2 | 69.55 | #317 |
| 14 | Qwen 3.5 35B A3B (Thinking) | 69.41 | #256 (Qwen 3.5 35B A3B) |
| 15 | Muse Glimmer 30B (High) | 68.93 | #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-sctpublic · How It Works · Data refreshed daily, snapshot 2026-10-20.