Medmarks - HEAD-QA v2 — leaderboard
Metric: Score (%). Source: medmarks.ai. 71 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Pro (Preview) | 93.97 |
| 2 | GPT-5.1 (Medium) | 92.33 |
| 3 | GPT-5.2 (Medium) | 91.95 |
| 4 | Grok 4 | 91.67 |
| 5 | Claude Sonnet 4.5 | 91.45 |
| 6 | GLM-4.7 FP8 | 91.32 |
| 7 | Qwen 3 235B A22B (Thinking) | 90.05 |
| 8 | Qwen 3 Next 80B A3B (Thinking) | 89.11 |
| 9 | GLM 4.5 Air | 88.46 |
| 10 | INTELLECT-3 | 88.3 |
| 11 | GPT-OSS-120B (High) | 88.24 |
| 12 | MiniMax-M2.1 | 88.02 |
| 13 | MiniMax-M2 | 87.81 |
| 14 | Qwen 3 VL 30B A3B (Thinking) | 87.5 |
| 15 | GPT-OSS-120B (Medium) | 87.4 |
Interactive version: theaggregate.ai/benchmark?slug=medmarks-head-qa-v2 · How the rankings work · Data refreshed daily, snapshot 2026-07-22.