Medmarks - Open-Ended: leaderboard
Metric: Weighted Mean Win Rate (%). Source: medmarks.ai. Saturation forecast: Around November 2027. 12 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.2 (Medium) | 63.89 |
| 2 | GPT-5.1 (Medium) | 62.44 |
| 3 | GPT-OSS-120B (High) | 55.07 |
| 4 | Qwen 3 235B A22B (Thinking) | 51.61 |
| 5 | Claude Sonnet 4.5 | 49.98 |
| 6 | Gemini 3 Pro (Preview) | 47.13 |
| 7 | GLM-4.7 FP8 | 45.19 |
| 8 | GPT-OSS-20B (High) | 42.66 |
| 9 | Qwen 3 30B A3B (Thinking) | 41.3 |
| 10 | Qwen 3 8B (Thinking) | 36.34 |
Interactive version: theaggregate.ai/benchmark?slug=medmarks-open-ended · How It Works · Data refreshed daily, snapshot 2026-10-09.