Medmarks - Medbullets OP4: leaderboard
The "Medbullets OP4" 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 | Gemini 3 Pro (Preview) | 90.91 | #66 |
| 2 | GPT-5.2 (Medium) | 90.04 | #110 (GPT-5.2) |
| 3 | GPT-5.1 (Medium) | 89.5 | #135 (GPT-5.1) |
| 4 | Qwen 3.6 27B (Thinking) | 87.99 | #170 (Qwen 3.6 27B) |
| 5 | GLM-4.7 FP8 | 87.99 | #209 |
| 6 | Grok 4 | 87.77 | #181 |
| 7 | Qwen 3.5 35B A3B (Thinking) | 87.77 | #256 (Qwen 3.5 35B A3B) |
| 8 | Qwen 3.5 27B (Thinking) | 87.45 | #187 (Qwen 3.5 27B) |
| 9 | Qwen 3.6 35B A3B (Thinking) | 86.9 | #226 (Qwen 3.6 35B A3B) |
| 10 | Muse Glimmer 30B (High) | 86.04 | #178 (Muse Glimmer 30B) |
| 11 | Qwen 3 235B A22B (Thinking) | 85.71 | #313 (Qwen 3 235B A22B) |
| 12 | GPT-OSS-120B (High) | 85.39 | #341 (GPT-OSS-120B) |
| 13 | Gemma 4 31B (IT) | 84.74 | #176 |
| 14 | Muse Glimmer 30B (xHigh) | 84.42 | #178 (Muse Glimmer 30B) |
| 15 | GPT-OSS-120B (Medium) | 84.31 | #341 (GPT-OSS-120B) |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=medmarks-medbullets-op4 · How It Works · Data refreshed daily, snapshot 2026-10-20.