MedSafe-Dx - Safety Pass Rate: leaderboard
Metric: Safety Pass Rate (%). Source: msdx.cortico.health. Saturation forecast: Estimated already saturated. 23 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-5.2 | 97.6 |
| 2 | Llama 4 Maverick | 96.8 |
| 3 | GPT-5.4 Mini | 96.4 |
| 4 | Claude Haiku 4.5 | 95.6 |
| 5 | Claude Sonnet 4.6 | 94.8 |
| 6 | GPT-5.6 Terra | 94.4 |
| 7 | GPT-5 Chat | 94 |
| 8 | GLM-5.3 | 93.2 |
| 9 | O3 Pro | 92.8 |
| 10 | GPT-5.6 Luna | 92 |
| 11 | GPT-5.6 Sol | 91.6 |
| 12 | Grok 4.6 | 90.8 |
| 13 | DeepSeek R1 | 90.4 |
| 14 | Grok 4.20 | 89.6 |
| 15 | GPT-5 Mini | 88.8 |
Interactive version: theaggregate.ai/benchmark?slug=medsafe-dx-safety-pass-rate · How It Works · Data refreshed daily, snapshot 2026-09-26.